Ontology matching: A literature review
Expert Systems with Applications 42 (2015) 949–971
Contents lists available at ScienceDirect
Expert Systems with Applications
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / e s w a
Ontology matching: A literature review
http://dx.doi.org/10.1016/j.eswa.2014.08.032
0957-4174/� 2014 Elsevier Ltd. All rights reserved.
⇑ Corresponding author.
E-mail addresses: locerdeira@uvigo.es (L. Otero-Cerdeira), franjrm@uvigo.es
(F.J. Rodríguez-Martínez), alma@uvigo.es (A. Gómez-Rodríguez).
Lorena Otero-Cerdeira ⇑, Francisco J. Rodríguez-Martínez, Alma Gómez-Rodríguez
LIA2 Research Group, Computer Science Department, University of Vigo, Spain
a r t i c l e i n f o a b s t r a c t
Article history:
Available online 30 August 2014
Keywords:
Ontology matching
Literature review
Classification framework
User survey
The amount of research papers published nowadays related to ontology matching is remarkable and we
believe that reflects the growing interest of the research community. However, for new practitioners that
approach the field, this amount of information might seem overwhelming. Therefore, the purpose of this
work is to help in guiding new practitioners get a general idea on the state of the field and to determine
possible research lines.
To do so, we first perform a literature review of the field in the last decade by means of an online
search. The articles retrieved are sorted using a classification framework that we propose, and the differ-
ent categories are revised and analyzed. The information in this review is extended and supported by the
results obtained by a survey that we have designed and conducted among the practitioners.
� 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Ontology matching is a complex process that helps in reducing
the semantic gap between different overlapping representations of
the same domain. The existence of such different representations
obeys to the natural human instinct to have different perspectives
and hence to model problems differently. When these domains are
represented using ontologies, the solution typically involves the
use of ontology matching techniques.
Ontologies and ontology matching techniques are an increasing
trend as ontologies provide probably the most interesting opportu-
nity to encode meaning of information. The last decades have born
witness to a period of extensive research in this field. Nowadays,
far from dying down, the activity seems to be increasing and
new publications, where the ontology matching field is addressed,
are continuously being released.
This reflects the global interest in ontology matching which we
have studied by means of an analytical review of the literature so
far. Other works and publications have successfully presented the
state-of-the-art in the field such as, Euzenat (2004), Shvaiko and
Euzenat (2013) and Kalfoglou and Schorlemmer (2003b), although
our purpose is quite different. We aim at retrieving articles related
to ontology matching that have been published in the last decade,
to classify and identify research lines relevant for ontology match-
ing. We also aim at providing a reference framework for the inte-
gration and classification of such articles. Therefore practitioners
approaching the field for the first time would be aware of the dif-
ferent types of publications regarding the field to better choose
those that better fits their needs, they would gain knowledge about
the main issues where the researchers have been working and
what are the main trends and challenges still to be addressed in
the next years.
To this end, the remainder of the paper is organized as follows.
In Section 2 a methodology to extract the articles is presented. Sec-
tion 3 presents general statistical results of the retrieved publica-
tions. Then, Section 4 illustrates the classification framework
proposed and describes each one of the categories defined. In Sec-
tion 5 we describe the limitations of the literature review and sug-
gest a practitioner-oriented survey to support the results of the
review. Such review is detailed in Section 6, and its limitations in
Section 7. Finally, in Section 8 we present our discussion, conclud-
ing remarks and directions for future work.
2. Procedures
To retrieve the articles for this literature review, several well-
known online databases were queried to obtain articles related
to the ontology matching field. As result over 1600 articles were
obtained which were filtered to narrow down the selection to
the final 694 articles that are included in this review. This screen-
ing allowed dismissing 58.09% of the initially retrieved articles.
Although this is a high percentage of the articles it is worth notic-
ing that the original search was broadly defined so an important
number of false positives among the initial results were already
expected. The screening of the articles was manually done, and
http://crossmark.crossref.org/dialog/?doi=10.1016/j.eswa.2014.08.032&domain=pdf
http://dx.doi.org/10.1016/j.eswa.2014.08.032
mailto:locerdeira@uvigo.es
mailto:franjrm@uvigo.es
mailto:alma@uvigo.es
http://dx.doi.org/10.1016/j.eswa.2014.08.032
http://www.sciencedirect.com/science/journal/09574174
http://www.elsevier.com/locate/eswa
950 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
took over 3 months to complete it due to the several iterations and
the amount of articles reviewed.
This review covers journal articles and conference proceedings
published within the last decade. Other publication forms such
as books, newspapers, doctoral dissertations, posters, etc., were
not considered as researchers use mainly journals and conference
papers to obtain and spread knowledge, thus these types of publi-
cations encompass the majority of the research papers published
about this subject.
The procedure followed to identify and filter the papers is
reflected in Fig. 1. First, different online databases were queried
using a combination of the following search strings: ontology
matching, ontology mapping and ontology alignment. These dat-
abases were: IEEEXplore Digital Library (IEEXplore, 2013), Science
Direct (Direct, 2013), ACM Digital Library (ACM, 2013) and Scopus
(Scopus, 2013).
In addition to these databases, the articles published in the
Ontology Alignment Evaluation Initiative (OAEI) (OAEI, 2013) were
also included as this initiative is considered as the most prominent
one regarding the evaluation of different matching systems and it
helps practitioners improve their works on matching techniques.
Sorting these data sources by amount of articles we have: Sco-
pus (626), ACM Digital Library (383), Science Direct (267), OAEI
(254) and IEEEXplore Digital Library (126), making this way a total
of 1656 articles initially obtained.
The next step was to dismiss those articles that were dupli-
cated, i.e, that have been obtained through two or more data
sources. When this situation arose, the criterion followed was to
dismiss the articles belonging to the data source with a higher
number of articles. By doing so, 335 articles were removed.
Next, the 1321 remaining articles were analyzed considering
their keywords and abstracts. Those whose keywords did not
include specific mentions to the ontology matching field or whose
abstracts did not introduce a research regarding this field were
excluded. Of all the criteria considered, this was the one that pro-
duced the sharpest cut down in the amount of articles. Addition-
ally, while reviewing the keywords and abstracts, also the papers
that corresponded to a poster publication were dismissed.
Finally, the 795 articles remaining were carefully reviewed to
dismiss those that did not consider ontology matching as their core
Fig. 1. Procedure followed to retrieve th
part. By applying this criterion another 101 articles were excluded,
therefore leaving the 694 articles that are included in this litera-
ture review.
3. Statistical results
In this section some statistical information about the articles is
presented and discussed. Articles were analyzed regarding publica-
tion year and database from which they were obtained.
3.1. Articles sorted by publication year
Fig. 2 represents the progression of the number of articles with
respect to their publication years. The measurements and values
that shape this progression are shown in Table 1. In this figure
we can observe that the amount of published articles steadily
increases from 2003 to 2012, where it peaks. The sharpest rise
was found between 2005 and 2006 where the percentage of pub-
lished articles rose from 3.75% to 8.36%. Between 2012 and 2013
we observe a pronounced decrease in the amount of published
articles, although as this review covers only the first semester of
2013, it is highly likely that the amount of published articles by
the end of the year would follow the increasing trend of the previ-
ous years. This increasing pattern reflects the global interest of the
research community in the ontology matching field.
3.2. Articles sorted by data source
To retrieve the articles for this literature review, a total of five
different data sources were used. Among these, four are well-
known online databases that were queried to obtain the articles.
The other source was the Ontology Alignment Evaluation Initiative
site, where all the publications related to this initiative are
published.
The classification of the retrieved articles by data source is
shown in Table 2 and graphically depicted in Fig. 3. These data
state that Scopus provides the highest amount of articles to the
total (41.79%, 290 articles) probably because it includes a wider
variety of source journals. On the other hand, ScienceDirect pro-
e articles for the literature review.
Fig. 2. Articles sorted by publication year.
Table 1
Articles with respect to publication year.
Publication year Number of articles Percentage over total (%)
2003 4 0.58
2004 17 2.45
2005 26 3.75
2006 58 8.36
2007 75 10.81
2008 84 12.10
2009 83 11.96
2010 88 12.68
2011 100 14.41
2012 107 15.42
2013 52 7.49
Total 694
Table 2
Articles with respect to data source.
Data source Number of
articles
Percentage over total
(%)
ACM Digital Library 72 10.37
IEEExplore Digital Library 114 16.43
Ontology Alignment Evaluation
Initiative
174 25.07
ScienceDirect 44 6.34
Scopus 290 41.79
Total 694
Fig. 3. Articles sorted by data source.
Fig. 4. Classification Framework.
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 951
vided significantly less articles than any of the other data sources
analyzed (6.34%, 44 articles). The remaining data sources consid-
ered respectively provide the 10.37% (72 articles, ACM Digital
Library), 16.43% (114, IEEExplore Digital Library) and 25.07% (174
articles, OAEI).
During the first filtering step (see Fig. 1) after initially retrieving
the articles from the different data sources, they were processed in
order to remove those duplicate results. In this situation the article
was always removed from the data source that had more articles as
the impact in the overall results for each database would be less
significative than in the case of dismissing those from the data
sources with less articles.
4. Classification
Relying on the analysis of the articles selected for the literature
review, we have defined an abstract framework that helps classify-
ing them. This framework, depicted in Fig. 4, shows six different
types of articles and it is in line with the general outline of the
main issues in ontology matching proposed by Euzenat and
Shvaiko in Euzenat and Shvaiko (2007, 2013). The different catego-
ries identified cover the most prominent fields of interest in ontol-
ogy matching.
� Reviews. This category includes the publications devoted to sur-
veying and reviewing the field of ontology matching. It also
includes those articles focused on detailing the state-of-the-
art as well as the future challenges in this field.
� Matching techniques. This category covers the publications
focused on different similarity measures, matching strategies
and methodologies, that can be used in the matching systems.
� Matching systems. This category includes those articles intro-
ducing new matching systems and algorithms, and also those
detailing enhancements to existing ones.
� Processing frameworks. This category comprehends the articles
that delve into the different uses of the alignments, i.e, those
operations that can be performed from alignments, such as
ontology merging, reasoning or mediation.
� Practical applications. This category covers those articles that
describe matching solutions applied to a real-life problem.
� Evaluation. This category covers the articles describing different
available approaches to evaluate the matching systems, as well
as the different existing benchmarks and the most relevant per-
formance measures.
The results of classifying the articles within the framework
defined are summarized in Table 3. According to our results the
Table 3
General results of the classification.
Category Number of articles Percentage (%)
Reviews 46 6.63
Matching techniques 85 12.25
Matching systems 302 43.52
Processing frameworks 147 21.18
Practical applications 76 10.95
Evaluation 38 5.48
Total 694
952 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
greater efforts have been focused on developing matching systems
and frameworks that exploit the alignments, while the evaluation
of such systems as well as the review of the field and the develop-
ment of applied solutions have not been object of such dedication.
In Table 4 the results of the classification are shown sorted by
year, identifying the amount of articles that match into each cate-
gory and its yearly percentage. These results are graphically com-
pared in Fig. 5 where the evolution can be more clearly
distinguished.
The amount of articles into Reviews showed an slight but
upward trend within the first five years of the time frame consid-
ered, peaking in 2008. Ever since, the trend in the amount of
reviews per year has remained almost constant. The evolution of
the Matching Techniques did not show any significant change until
2007 when the amount of articles rose from three to thirteen. After
this sharp increase, the values remained constant for the next year
and then they fell to six in 2009. In 2010 there was another signif-
icant increase, followed by yet another fall in the amount of arti-
cles. Apparently the periods of intense work in this category are
followed by others where the amount of publications is signifi-
cantly lower.
Regarding the Matching Systems, the amount of published arti-
cles shows an upward trend in the considered period, since, as sta-
ted before, the values for 2013 can not be considered definitive as
the articles for this review were retrieved in the first semester of
the year. Regardless of the general upward trend, the amount of
articles in this category reached some local peaks in 2006, 2009
and 2012, followed by periods of less activity. Anyway, it is impor-
tant to notice that the periods of lower activity in this category cor-
respond with periods of higher activity in matching techniques and
vice versa. It is highly likely that the same researchers that define
new matching systems are the ones that had suggested new
matching techniques to be used within them, which also validates
the process of construction of a matching system.
The evolution in the number of Processing Frameworks was
increasingly constant until 2007 when it reached a local peak, just
to show a slight downward trend for two years, before starting a
continuos rise that took the amount of published articles regarding
this category to its top value in 2012.
Regarding the Practical applications, the amount of articles
shows in general an upward trend. It reaches a local peak in
2007 with 8 articles. In 2008 it shows a slight decrease. However,
since 2010, the number of articles continues to rise every year. It is
worth noticing that even when the data for 2013 does not cover
the whole year, the amount of articles devoted to practical applica-
tions was already a 66.6% of the articles published in 2012 in the
same category.
Finally, the publications related to Evaluation show an evolution
pattern really similar to the one detected for the reviews, not show-
ing any significant behavior.
After providing some general outline of the evolution of the dif-
ferent categories over the years, in the following sections a deeper
analysis of each one of them is included, considering inner classifi-
cations for each one of these general categories.
4.1. ‘Reviews’ category
Within this category the publications have been further sorted
according to their scope, namely, the publications were identified
as being of either general purpose or specific purpose, as depicted
in Fig. 6. More than half of the 46 articles included in this category,
26, were considered general purpose reviews as they offer some
insight into the ontology matching field without specifically
emphasizing on any subject. Among these, there are for instance,
surveys (Falconer, Noy, & Storey, 2007; Shvaiko & Euzenat, 2005;
Thayasivam, Chaudhari, & Doshi, 2012; Zhu, 2012), state-of-the-
art articles (Droge, 2010; Gal & Shvaiko, 2008; Ngo, Bellahsene, &
Todorov, 2013) and publications unveiling the future challenges of
the field (Kotis & Lanzenberger, 2008; Shvaiko & Euzenat, 2013).
In turn, the remaining 20 articles show a more limited scope
within the field. When analyzing these articles we have stated that
they were devoted either to delving into a very specific area within
the ontology matching field or to studying the feasibility of apply-
ing ontology matching to a certain domain.
Some of the specific fields within ontology matching that were
addressed cover topics such as (i) matching across different lan-
guages (Fu, Brennan, & O’Sullivan, 2009), (ii) instance-based match-
ing (Castano, Ferrara, Lorusso, & Montanelli, 2008), which is one
among the different types of techniques to perform ontology
matching. Other articles developed the subject of (iii) external
sources for ontology matching (Fugazza & Vaccari, 2011; Lin &
Sandkuhl, 2008a), these techniques take advantage of auxiliary or
external resources in order to find matchings to terms based on lin-
guistic relations between them such as synonymy or hyponymy.
These external resources are usually lexicons or thesauri.
On the other hand, among the domains where ontology match-
ing could be used, we have identified articles on domains such as
geography (Tomaszewski & Holden, 2012), medicine (Wennerberg,
2009) or agriculture (Lauser et al., 2008).
4.2. ‘Matching Techniques’ category
Ontology matching techniques propose different approaches
for the matching that are implemented in ontology matching
algorithms. When building an ontology matching system, differ-
ent algorithms are usually used, exploiting therefore different
ontology matching techniques. In this category two different
types of articles have been identified. Some articles are devoted
to describing new or enhanced similarity measures and to analyz-
ing the building blocks of the ontology matching algorithms, while
others make use of such artifacts to define matching strategies or
methodologies.
In total there are 85 articles in this category, where 57:65% (49
articles) belong to the first group of basic matching techniques and
42:35% (36 articles) belong to complex matching techniques.
The different matching techniques have been subject of study in
the latest years. For the purpose of this review, to sort the (i) basic
matching techniques we have followed the classification proposed
by Euzenat and Shvaiko (Euzenat & Shvaiko, 2013), depicted in
Fig. 7, since to the best our knowledge is the most complete one
and reflects most of the other previous classifications. This classifi-
cation is an evolution of another previously proposed by the same
authors in Euzenat and Shvaiko (2007).
This classification can be followed top-down and therefore
focusing on the interpretation that the different techniques offer
to the input information, but also bottom-up, focusing on the type
of the input that the matching techniques use. Despite the fol-
lowed approach both meet at the concrete techniques tier.
Following the top-down interpretation, the matching tech-
niques can be classified in a first level as:
Table 4
Results of the classification.
Year Category Number of articles Total Annual percentage
2003 Reviews 1 4 25.00%
Matching techniques – –
Matching systems 1 25.00%
Processing frameworks 2 50.00%
Practical applications – –
Evaluation 0 –
2004 Reviews 1 17 5.88%
Matching techniques 5 29.41%
Matching systems 6 35.29%
Processing frameworks 3 17.65%
Practical applications – –
Evaluation 2 11.76%
2005 Reviews 4 26 15.38%
Matching techniques 3 11.54%
Matching systems 6 23.08%
Processing frameworks 9 34.62%
Practical applications 1 3.85%
Evaluation 3 11.54%
2006 Reviews 2 58 3.45%
Matching techniques 3 5.17%
Matching systems 30 51.72%
Processing frameworks 18 31.03%
Practical applications 4 6.90%
Evaluation 1 1.72%
2007 Reviews 4 75 5.33%
Matching techniques 13 17.33%
Matching systems 27 36.00%
Processing frameworks 19 25.33%
Practical applications 8 10.67%
Evaluation 4 5.33%
2008 Reviews 8 84 9.52%
Matching techniques 13 15.48%
Matching systems 34 40.48%
Processing frameworks 16 19.05%
Practical applications 5 5.95%
Evaluation 8 9.52%
2009 Reviews 5 83 6.02%
Matching techniques 6 7.23%
Matching systems 46 55.42%
Processing frameworks 11 13.25%
Practical applications 10 12.05%
Evaluation 5 6.02%
2010 Reviews 3 88 3.41%
Matching techniques 18 20.45%
Matching systems 37 42.05%
Processing frameworks 16 18.18%
Practical applications 10 11.36%
Evaluation 4 4.55%
2011 Reviews 4 100 4.00%
Matching techniques 9 9.00%
Matching systems 47 47.00%
Processing frameworks 21 21.00%
Practical applications 13 13.00%
Evaluation 6 6.00%
2012 Reviews 6 107 5.61%
Matching techniques 9 8.41%
Matching systems 52 48.60%
Processing frameworks 23 21.50%
Practical applications 15 14.02%
Evaluation 2 1.87%
2013 Reviews 8 52 15.38%
Matching techniques 6 11.54%
Matching systems 16 30.77%
Processing frameworks 9 17.31%
Practical applications 10 19.23%
Evaluation 3 5.77%
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 953
Fig. 5. Evolution of the different categories.
Fig. 6. Specific types of articles within ‘Reviews’ category.
954 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
– Element-level matchers: these techniques obtain the corre-
spondences by considering the entities in the ontologies in
isolation, therefore ignoring that they are part of the struc-
ture of the ontology.
– Structure-level matchers: these techniques obtain the corre-
spondences by analyzing how the entities fit in the structure
of the ontology.
In the second level of the classification, the techniques can be
further classified as:
– Syntactic: these techniques limit their input interpretation to
the instructions stated in their corresponding algorithms.
– Semantic: these techniques use some formal semantics to
interpret their input and justify their results.
If reading the classification bottom-up, the elementary match-
ing techniques can be initially divided in two categories deter-
mined by the origin of the information considered for the
matching process:
– Content-based: these techniques focus on the internal infor-
mation coming from the ontologies to be matched.
– Context-based: these techniques consider for the matching,
external information that may come from relations between
ontologies or other external resources (context).
In the second level of the classification, both categories are fur-
ther refined. The content-based category is further divided into four
new groups, depending on the input that the techniques use:
– Terminological: these methods consider their inputs as
strings.
– Structural: these methods are based on the structure of the
entities (classes, individuals, relations) found in the
ontology.
– Extensional: these methods compute the correspondences by
analyzing the set of instances of the classes (extension).
– Semantic: these techniques need some semantic interpreta-
tion of the input and usually use a reasoner to deduce the
correspondences.
In the second level of the classification for the context-based cat-
egory, the techniques can be also further classified as syntactic or
semantic techniques.
The next level in any of both classifications already corresponds
to the specific techniques. Following the different paths in this
classification tree, several techniques may be reached. These cate-
gories were used to further sort the 49 articles belonging to basic
matching techniques. The classification of these articles was partic-
ularly hard since most of them were not devoted to a single tech-
nique but to several, so in the following we provide an example of
articles that match into some categories but it is worth noticing
that many of them may also be included in another one.
� Formal Resource-based: these techniques use formal resources to
support the matching process, such as upper level ontologies,
domain-specific ontologies or the recorded alignments of previ-
Fig. 7. Matching techniques classification. Extracted from the book ‘Ontology Matching’ (Euzenat & Shvaiko, 2013).
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 955
ously matched ontologies (alignment reuse). Examples of such
techniques are Scharffe, Zamazal, and Fensel (2013) and
Mascardi, Locoro, and Rosso (2010).
� Informal Resource-based: these techniques, as those in the previ-
ous category, also exploit an external resource, but in this case
the external resources are informal ones. This group of tech-
niques deduce relations between ontologies using the relation
between the ontologies and such informal resources. An exam-
ple of such category could not be found among the results of the
review.
� String-based: these techniques are based on the similarity of the
strings that represent the names and descriptions of the entities
in the ontologies. There are several string distance metrics that
can be used in these methods Levenshtein, Jaccard, Jaro-Winkler,
Euclidean, TFIDF, etc., (Cohen, Ravikumar, & Fienberg, 2003).
Such techniques are present, for instance in the work of
Akbari, Fathian, and Badie (2009).
� Language-based: these techniques rely on Natural Language Pro-
cessing, as these do not consider names as simply strings but
words in some natural language. Techniques in this category
are, for example, tokenisation, lemmatisation or stopword elimi-
nation, some of which are applied by Shah and Syeda-Mahmood
in Shah and Syeda-Mahmood (2004). This category also consid-
ers those techniques that take advantage from external
resources to find similarities between terms, using for instance,
lexicons, dictionaries or thesauri. In He, Yang, and Huang (2011)
for instance, the WordNet (WordNet, 2013) database is used as
the external resource.
� Constraint-based: these techniques consider criteria regarding
the internal structure of the entities, such as the domain and
range of the properties or the types of the attributes, to calculate
the similarity between them. It is common to use these tech-
niques in combination with others as in the work by
Glückstad (2010).
� Graph-based: these techniques consider the ontologies to match
as labelled graphs, or even trees, and treat the ontology match-
ing problem as a graph homomorphism problem. An example of
these techniques can be found in the paper from Joslyn, Paulson,
and White (2009). This category also considers those techniques
that exploit as external resources, repositories where ontologies
and their fragments, together with certain similarity measures
are stored. A proposal in this line can be found in the paper from
Aleksovski, Ten Kate, and Van Harmelen (2008).
� Taxonomy-based: these techniques can be seen as a particular
case of the previous ones which only consider the specialization
relation. Examples were these techniques were applied could
not be found in the articles belonging to basic matching tech-
niques. However, an example of its application can be found in
the work by Warin and Volk (2004).
� Instance-based: these techniques exploit the extension of the
classes in the ontologies, i.e., the individuals, with the intuition
that if the individuals are alike, then the classes they belong to
should also be similar. These techniques can use set-theoretic
principles but also more elaborated statistical techniques. In
this category we can classify the work by Loia, Fenza, De Maio
and Salerno presented in Loia, Fenza, De Maio, and Salerno
(2013).
� Model-based: these techniques exploit the semantic interpreta-
tion linked to the input ontologies. An example of this category
are the description logics reasoning techniques, which are applied
in the work published by Sánchez-Ruiz, Ontañón, González-
Calero, and Plaza (2011).
As mentioned at the beginning of this section, the techniques
we have just presented, are the building blocks upon which (ii)
complex matching techniques are built. This category includes meth-
odologies that propose different ways to tackle the matching prob-
lem but from a higher point of view. Examples of this can be found
in Cohen et al. (2003) where Cohen, Ravikumar and Fienberg pro-
pose the partitioning of the ontologies before starting the matching
process, in Dargham and Fares (2008) where the Dargham and
Fares present their methodology which takes as basis some well-
known algorithms or in Acampora, Loia, Salerno, and Vitiello
(2012) where Acampora, Loia, Salerno and Vitiello propose the
use of a memetic algorithm to perform the alignment between
two ontologies.
956 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
Other matching techniques also included in this category are
those that can not be considered basic or that take advantage of other
aspects of building a matching solution that are not directly related
to computing the alignments. There are for instance articles devoted
to presenting (i) different ways of aggregating the results from differ-
ent similarity measures (Lai et al., 2010; Lin & Sandkuhl, 2007; Tian &
Guo, 2010), others that (ii) combine the results of different matchers
(Liu et al., 2012). There are others that also include techniques inher-
ited from other fields such as (iii) learning methods (Rubiolo,
Caliusco, Stegmayer, Coronel, & Fabrizi, 2012; Todorov, Geibel, &
Kühnberger, 2010), probabilistic methods (Calı̀, Lukasiewicz,
Predoiu, & Stuckenschmidt, 2008; Spiliopoulos, Vouros, &
Karkaletsis, 2010) or those that consider the user’s involvement
(Lin & Sandkuhl, 2008b) in the matching process.
Despite the various types of matching techniques that have
been presented in this section, both basic and complex ones, we
are certain that many others will continue to arise. Some of these
may offer a revisited version of previously exiting techniques,
but more likely new types of techniques will be defined specially
for those categories not fully explored yet. The main challenges
these techniques face is their efficiency (Kotis & Lanzenberger,
2008; Shvaiko & Euzenat, 2013). Most of them describe approaches
to matching ontologies which at a theoretical level may obtain a
positive outcome. However, depending on the type of application
where these techniques will be implemented, not all techniques
will be equally valid. For instance, if considering a dynamic appli-
cation, the amount of resources employed in terms of memory and
time consumption should be kept to a minimum.
4.3. ‘Matching Systems’ category
This category contains the articles focused on detailing new
matching algorithms and systems as well as enhancements, mod-
ifications or different approaches to previously defined ones. Some
of these systems are well-known in the research community as
they have participated for several years in the Ontology Alignment
Evaluation Initiative. Although the purpose of this review is not to
compile an exhaustive list of all the existing systems, it is worth
mentioning some of the most relevant ones, such as:
� AgreementMaker (Cruz, Antonelli, & Stroe, 2009a) is a schema
and ontology matching system. It allows a high customization
of the matching process, including several matching methods
to be run on inputs with different levels of granularity, also
allowing to define the amount of user participation and the for-
mats that the input ontologies as well as the results of the align-
ment may be stored in. This system has a high level of maturity
because from 2007 there is a continuous flow of publications
describing its foundations and enhancements: Sunna and Cruz
(2007), Cruz, Antonelli, Stroe, Keles, and Maduko (2008), Cruz,
Antonelli, and Stroe (2009b, 2009c), Cruz et al. (2010),
Pesquita, Stroe, Cruz, and Couto (2010), Cross et al. (2011),
Cruz, Stroe, Pesquita, Couto, and Cross (2011) and Cruz et al.
(2011).
� Anchor-Flood (Hanif & Aono, 2009) is an algorithm for ontology
matching that starts out of an initial anchor, i.e, a pair of alike
concepts between the ontologies. From this anchor using neigh-
borhood concepts, new anchors are identified and the algorithm
continues. Anchor-Flood was developed from 2008 to 2009,
having in this period a total of 3 reported papers, (Hanif &
Aono, 2008a, 2008b, 2009). This system has been tested in
two different campaigns of OAEI.
� AOAS (Zhang & Bodenreider, 2007) is an ontology matching sys-
tem specifically devoted to aligning anatomical ontologies. It
takes as input OWL ontologies and identifies 1:1, 1:n and n:m
alignments. It is a hybrid approach that uses both direct tech-
niques, such as lexical and structural ones, and indirect tech-
niques, that consist in the identification of correspondences
by means of a reference ontology. AOAS is a really specific sys-
tem which, in the period from 2003 to 2013, only accounts for
one publication, (Zhang & Bodenreider, 2007).
� AROMA (David, 2011) finds equivalence and subsumption rela-
tions between classes and properties of two different taxono-
mies. It is defined as an hybrid, extensional and asymmetric
approach that lays its foundations on the association rule para-
digm and statistical measures. AROMA’s developers have been
constantly working on it since 2006. There has been at least
one paper published every year detailing this system and its
evolution for the past 7 years. (David, Guillet, & Briand, 2006;
David, 2007, 2008b, 2008a, 2011). This system has taken part
in several editions of the OAEI.
� ASCO (Le, Dieng-Kuntz, & Gandon, 2004) exploits all the infor-
mation available from the entities in the ontologies, names,
labels, descriptions, information about the structure, etc, to
compute two types of similarities, a linguistic and a structural
one, which are lately combined. The development of ASCO
started in 2004, however it was discontinued until 2007 when
it was reprised. In this 3 year-span, the publications describing
it are: Le et al. (2004) and Thanh Le and Dieng-Kuntz (2007).
� ASE (Kotis, Katasonov, & Leino, 2012a) is an automated ontology
alignment tool based on AUTOMSv2 that computes equivalence
and subsumption relations between two input ontologies. This
system was released in 2012 and tested that year’s edition of
the OAEI. So far, we have not found any other publication apart
from Kotis et al. (2012a), describing it. However it is possible
that other enhancements may be developed as this system
was already based on a previous system by the same authors
for which there was also a significant interval between the first
release and the subsequent updates.
� ASMOV (Behkamal, Naghibzadeh, & Moghadam, 2010) is an
algorithm that derives an alignment from the lexical and struc-
tural information of two input ontologies by computing a sim-
ilarity measure between them. This algorithm also includes a
step of semantic verification where the alignments are checked
so that the final output does not contain semantic inconsisten-
cies. The greatest efforts in maintaining ASMOV were concen-
trated in 2008 when authors published Jean-Mary,
Shironoshita, and Kabuka (2008) and Jean-Mary and Kabuka
(2008), however, this system was constantly maintained from
2007 to 2010. In this period the following articles describe its
features and performance: Jean-Mary and Kabuka (2007),
Jean-Mary, Shironoshita, and Kabuka (2009, 2010) and
Behkamal et al. (2010). Some of these articles detail the results
obtained by ASMOV in the different editions it took part in.
� AUTOMSv2 (Kotis, Katasonov, & Leino, 2012b) is an automated
ontology matching tool that was build as an evolution of the
previous tool AUTOMS (Kotis, Valarakos, & Vouros, 2006) which
was enhanced with more alignment methods and synthesizing
approaches as well as with multilingual support. Articles
describing AUTOMSv2 were published in 2008 and 2012. This
system shows one of the highest intervals of inactivity of the
studied ones. Due to the short time interval from the last article,
new articles describing AUTOMSv2 participation in OAEI or new
versions could be published.
� Coincidence-Based Weighting (Qazvinian, Abolhassani, (Hossein),
& Hariri, 2008) uses an evolutionary approach for ontology
matching. It takes as input OWL ontologies, which are pro-
cessed as graphs, and a similarity matrix between the concepts
of the ontologies, obtained from a string distance measure. It
includes a genetic algorithm to iteratively refine the mappings.
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 957
This system was developed for 2 years, since 2007 to 2008, as
reflect the articles describing it, (Haeri, Abolhassani,
Qazvinian, & Hariri, 2007; Qazvinian et al., 2008).
� CIDER (Gracia, Bernad, & Mena, 2011) is an ontology matching
system that extracts the ontological context of the compared
terms by using synonyms, hyponyms, domains, etc., and then
enriches it by means of some lightweight inference rules. This
system was developed using the Alignment API (David,
Euzenat, Scharffe, & dos Santos, 2011). This system has been
developed and maintained since 2008 to 2013. From the first
paper describing it (Gracia & Mena, 2008) until the next one
(Gracia et al., 2011), 3 years passed. Then again, publications
regarding this system, were interrupted until last year with
(Gracia & Asooja, 2013). This kind of systems developed over
several years are usually the product of a deep effort of the
developers to correct the issues detected in previous versions,
and account for the time-span existing between the publication
of the different articles.
� CODI (Huber, Sztyler, Nößner, & Meilicke, 2011) uses some lex-
ical similarity measures combined with schema information to
output the alignments between concepts, properties and indi-
viduals. It is based on the syntax and semantics of Markov logic,
and turns every matching problem into an optimization prob-
lem. According to the data in our review, CODI was developed
and maintained from 2010 to 2011. In these years it took part
in the corresponding OAEI contests (Huber et al., 2011;
Noessner & Niepert, 2010).
� COMA (Maßmann, Raunich, Aumüller, Arnold, & Rahm, 2011),
COMA++ (Nasir & Noor, 2010) are ontology matching systems
that have been developed over the last decade. They are highly
evolved and customizable systems that support the combina-
tion of different matching algorithms. These are highly evolved
systems since they have been maintained and updated until
2011. In this period 5 articles were found, devoted to describing
these systems, (Aumüller, Do, Massmann, & Rahm, 2005;
Aumueller, Do, Massmann, & Rahm, 2005; Do & Rahm, 2002;
Do, 2006; Engmann & Maßmann, 2007; Massmann, Engmann,
& Rahm, 2006; Maßmann et al., 2011; Nasir & Noor, 2010).
� DSSim (Nagy, Vargas-Vera, & Stolarski, 2009) is an ontology
matching system which combines the similarity values pro-
vided by both syntactic and semantic similarity algorithms to
then refine the correctness of the outputs by means of a belief
function. DSSim was developed from 2006 to 2009. In these
4 years, there was at least an article published every year detail-
ing the system, its continuous enhancements, and its results in
OAEI (Nagy, Vargas-Vera, & Motta, 2006, 2007; Nagy, Vargas-
Vera, Stolarski, & Motta, 2008; Nagy et al., 2009).
� Eff2Match (Chua & Kim, 2010) is an ontology matching tool that
follows a process of anchor generation and expansion. This sys-
tem is particularly focused on achieving an efficient perfor-
mance and therefore it includes several techniques to reduce
the amount of possible candidates to avoid unnecessary com-
parisons. Considering the data obtained from our literature
review, Eff2Match only has a related publication that was
released in 2010 that presents its results for OAEI’10.
� FalconAO (Hu & Qu, 2008) obtains the alignment between the
input ontologies by internally running two algorithms, a lin-
guistic one (LMO) (Zhang, Hu, & Qu, 2011) as a first step, to then
use the alignments provided as an external output for the graph
matching algorithm (GMO) (Hu, Jian, Qu, & Wang, 2005) subse-
quently run. This system was continuously developed from
2005 to 2008 (Hu, Cheng, Zheng, Zhong, & Qu, 2006; Hu et al.,
2007; Hu & Qu, 2008; Jian, Hu, Cheng, & Qu, 2005). In 2010, a
new publication was released (Hu, Chen, Cheng, & Qu, 2010)
with the results of this system in the OAEI.
� FBEM (Stoermer & Rassadko, 2009a) is a matching system that is
mainly focused on instance matching, this approach considers
not only the similarity of entity features as keys and values,
but also the fact that some features are more relevant for iden-
tifying an entity than others. According to the data obtained in
this review, this system is described in just 2 publications
released in 2009 and 2010 respectively, (Stoermer & Rassadko,
2009b; Stoermer, Rassadko, & Vaidya, 2010).
� FuzzyAlign (Fernández, Velasco, Marsa-Maestre, & Lopez-
Carmona, 2012) is a fuzzy, rule-based ontology matching sys-
tem that outputs the alignments between two input ontologies
by exploiting the lexical and semantical information of the enti-
ties’ names and the inner structure of the ontologies. This sys-
tem was developed from 2009 to 2012, however, it was not a
constant maintenance, since in this period only 2 articles
account for its updates and enhancements, (Fernández,
Velasco, & López-Carmona, 2009; Fernández et al., 2012).
� GeRoMeSuite (Quix, Gal, Sagi, & Kensche, 2010) allows the
matching of models represented in different languages, for
instance XML Schemas with OWL ontologies. Besides it is a cus-
tomizable system that includes several matching algorithms
that may be combined according to different ways of aggrega-
tion and filtering. GeRoMeSuite is a mature system that has been
developed from 2007 to 2010 (Kensche, Quix, Li, & Li, 2007;
Quix, Geisler, Kensche, & Li, 2008, 2009; Quix et al., 2010). How-
ever, from 2010 to 2013 no new improvements were publicly
released. It took part in OAEI editions from 2008 to 2010.
� GLUE (Doan, Madhavan, Domingos, & Halevy, 2004) is a semi-
automatic ontology matching system that uses a set of com-
bined machine learning techniques to output the alignment
between the taxonomies of two input ontologies. This system
only has a publication released in 2004, which made us believe
that its development has been discontinued.
� GOMMA (Hartung, Kolb, Groß, & Rahm, 2013) uses several
matchers to evaluate both the lexical and structural similarity
of the entities from the input ontologies. It uses some enhanced
comparison techniques to compute parallel string matching on
graphical processing units. GOMMA was first released in 2011,
and it has continued to be maintained up to date. The maturity
level of this system is significative, as at least one publication
regarding its results has been released every year since it was
first developed (Groß, Hartung, Kirsten, & Rahm, 2012;
Hartung et al., 2013; Kirsten, Gross, Hartung, & Rahm, 2011).
This system has been tested so far in a edition of the OAEI,
(Groß et al., 2012).
� HCONE (Kotis & Vouros, 2004) is an approach to ontology merg-
ing that uses WordNet (WordNet, 2013), a lexical database, as
an external resource to obtain possible interpretations of the
concepts being matched. HCONE was an stable system main-
tained from 2004 to 2006 (Kotis & Vouros, 2004; Kotis,
Vouros, & Stergiou, 2006; Vouros & Kotis, 2005). After 2006
we could not retrieve any publication were it was used or
modified.
� Hertuda (Hertling, 2012) is a very simple string matcher that
separately handles the alignment of classes and properties,
and select among the possible ones those reaching certain
pre-established thresholds. Hertuda is a relatively young sys-
tem which was released in 2012. Within the limits of this liter-
ature review only two articles were found that described its
overall behavior and results in OAEI: Hertling (2012) and Grau
et al. (2013).
� HotMatch (Dang et al., 2012a) combines several matching strat-
egies, that exploit both the lexical and structural information, to
obtain the alignments between the ontologies. Some filters are
included to remove the false-positive mappings from the final
958 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
output. As happens with Hertuda, HotMatch is also a young sys-
tem developed from 2012 up to date (Dang et al., 2012b; Grau
et al., 2013), hence, new versions and enhancements are to be
expected. This system, has also taken part in OAEI’12.
� HMatch (Castano, Ferrara, & Messa, 2006) is an ontology match-
ing system that linearly combines a linguistic affinity value and a
contextual affinity one to compute the similarity of concept
names and contexts. Internally it codifies the ontologies as
graphs. HMatch was developed from 2006 to 2009 (Castano
et al., 2006; Castano, Ferrara, Lorusso, & Montanelli, 2007;
Castano et al., 2008). According to the articles retrieved for this
review, HMatch only took part in the edition of 2006 of the OAEI.
� IF-MAP (Kalfoglou & Schorlemmer, 2003a) is a matching system
that lies its foundations on the mathematical theory of semantic
information flow. To match two input ontologies it uses a refer-
ence ontology used as common reference. In the time-span con-
sidered in this literature review, only one article was found
devoted to this system. Such publication was released in
2003, it is possible that there are other articles that had pub-
lished before that, which would fall outside the scope of this
review, anyhow, considering the time elapsed, we are prone
to believing that this system has been discontinued.
� iMatch (Albagli, Ben-Eliyahu-Zohary, & Shimony, 2012) is a
probabilistic ontology matching system based on Markov net-
works. It takes OWL ontologies as input and outputs 1:1 align-
ments. The matching is tackled as a graph matching problem
where the initial similarity between the nodes is provided, for
instance, by the users. iMatch was first released in 2009
(Albagli, Ben-Eliyahu-Zohary, & Shimony, 2009) and then revis-
ited in 2012 (Albagli et al., 2012).
� KOSImap (Reul & Pan, 2010) uses description logic reasoning to
firstly obtain implicit background knowledge for every entity in
the ontologies, then build a similarity matrix out of the three
types of similarities computed for the identified pairs of enti-
ties, and finally dismiss those mappings considered false-posi-
tives. KOSImap was only maintained for 2 years, since 2009,
when it took part in the OAEI, (Reul & Pan, 2009) to 2010
(Reul & Pan, 2010).
� LDOA (Kachroudi, Moussa, Zghal, & Yahia, 2011) combines some
well-known terminological and structural similarity measures,
but it also exploits an external resource by using Linked Data
which provides additional information to the entities being
matched. In the interval considered in this literature review,
we have only retrieved one article devoted to describing this
system, in 2011, describing its behavior in the OAEI. As it is rel-
atively contemporary, new enhancements and publications are
still to be expected.
� Lily (Wang, 2011) combines different matching strategies to
adapt itself to the problem being tackled at each moment, gen-
eric ontology matching (GOM) for the normal-sized ontologies
and large scale ontology matching (LOM) for more demanding
matching tasks. It also includes a mapping debugging function
used to improve the alignment results and to dismiss the faulty
ones. Lily was first released in 2007 (Wang & Xu, 2007) and con-
tinued to take part in the OAEI contests until 2011 (Wang & Xu,
2008; Wang & Xu, 2009; Wang, 2011). We consider that the
reliability of this system has been enough proven as show the
different results obtained in the OAEI.
� LogMap (Jiménez-Ruiz & Cuenca Grau, 2011) is an ontology
matching iterative process that starting with a set on anchor
mappings obtained from lexical comparison, alternatively com-
putes mapping repair and mapping discovery steps. To discover
the new anchors structural information is also exploited. This
system has a high level of maturity as in the last 3 years there
have been at least 6 publication describing its performance
and results in the OAEI contests (Jiménez-Ruiz & Cuenca Grau,
2011; Jiménez-Ruiz, Grau, & Zhou, 2011; Jiménez-Ruiz,
Morant, & Grau, 2011; Jiménez-Ruiz, Grau, & Horrocks, 2012;
Jiménez-Ruiz, Meilicke, Grau, & Horrocks, 2013). In this period
LogMap’s developers have already implemented a light version
of LogMap, LogMaplt. We are prone to believing that developers
of this system will continue to improve it and include further
functionalities and versions.
� MaasMatch (Schadd & Roos, 2012a) computes a similarity cube
between the concepts in the ontologies which is the result of
aggregating a syntactic, a structural, a lexical and a virtual docu-
ment similarity. An extraction algorithm is run to dismiss the
faulty alignments from the final output. This system has been
constantly updated since it was first released in 2011, its reliabil-
ity and usefulness have been tested over the years by its partici-
pation in the OAEI (Schadd & Roos, 2011, 2012a, 2012b, 2013).
� MapPSO (Bock, Dänschel, & Stumpp, 2011) applies the particle
swam optimization technique (PSO) to compute the alignment
between two input ontologies. The MapEVO (Bock et al., 2011)
system, developed by the same authors, relies on the use of evo-
lutionary programming, another variant of population-based
optimization algorithms. MapPSO has been described in at least
5 different publications between 2008 and 2011. Among those
systems revised, MapPSO is one of those with the higher amount
of publications (Bock & Hettenhausen, 2008, 2010; Bock, Liu, &
Hettenhausen, 2009; Bock, 2010; Bock, Lenk, & Dänschel,
2010; Bock et al., 2011). These publications include the partici-
pation of MapPSO in editions of OAEI from 2008 to 2011.
� MapSSS (Cheatham, 2011) computes subsequently three types
of metrics, syntactic, semantic and structural, and any positive
result from any of them is included as a positive solution, and
then it explores the neighborhood of the newly matched pair
for new possible matches. Instead of defining a filtering system
which would dismiss possible pair after being selected, this sys-
tem works the other way round only selecting those nodes that
match to only another node, and therefore not risking the pos-
sibility of choosing a wrong solution. This system was released
in 2011, and, ever since it took part in the annual contest of the
OAEI. This system has been therefore significantly tested and
evaluated (Cheatham, 2011; Cheatham & Hitzler, 2013).
� MEDLEY (Hassen, 2012) is an ontology alignment system that
uses lexical and structural methods to compute the alignment
between classes, properties and instances. It also uses an exter-
nal dictionary to tackle the problem of having concepts
expressed in different natural languages. This system was
described in 2012 in Hassen (2012) where the results of its par-
ticipation in that year’s OAEI are summarized. Other publica-
tions, as well as its participation in new OAEI editions, are to
be expected because this system quite recent.
� MoTo (Fanizzi, d’Amato, & Esposito, 2011) takes OWL ontologies
as input and obtains equivalence relations between concepts. It
initially uses several matchers whose results are combined by
means of a metalearner. The alignments obtained are sorted,
discarding those invalid ones. The remaining are divided into
certain and uncertain. For the uncertain ones a validation pro-
cess is started aiming at recovering them. In this literature
review we have found publications describing this system both
in 2010 (Esposito, Fanizzi, & d’Amato, 2010) and 2011 (Fanizzi
et al., 2011), but so far, no other publication has been released.
� OACAS (Zghal, Kachroudi, Yahia, & Nguifo, 2011) is an algorithm
to align OWL-DL ontologies. It firstly transforms the ontologies
into graphs and then it combines and aggregates different sim-
ilarity measures. At each moment the most suitable similarity
measure is applied according to the type of the entities being
matched. It also exploits the neighboring relations of the enti-
ties. The first article describing OACAS was published in 2009
(Zghal, Kachroudi, Yahia, & Nguifo, 2009), then until 2011 no
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 959
new articles were retrieved (Zghal et al., 2011). Therefore, con-
sidering the two-year span between both, it is possible that new
articles will be published within this or the next year. This sys-
tem was publicly tested in the OAEI’11.
� OLA (Kengue, Euzenat, & Valtchev, 2007) performs the alignment
between two graph-represented ontologies and offers some
extended features to manipulate the output alignment. This sys-
tem was developed between 2004 (Euzenat & Valtchev, 2004)
and 2005 (Euzenat, Guégan, & Valtchev, 2005), later, it was
revisited in 2007 (Kengue et al., 2007). In this interval, it took
part in the OAEI of 2005 and 2007. However, so far, it has been
6 years where no additional articles regarding OLA were retrieve
by means of the queries run for this literature review.
� oMap (Straccia & Troncy, 2005c) automatically aligns two OWL
ontologies by using the prediction of different classifiers, such
as terminological, machine learning-based, or some based on
the structure and semantics of the OWL axioms. This system
was deeply revised in 2005 (Straccia & Troncy, 2005a; Straccia
& Troncy, 2005c; Straccia & Troncy, 2005b) when it was
released and then in 2006 (Straccia & Troncy, 2006). In 2005
it took part in the OAEI, however, ever since, no new articles
describing oMap have been published.
� OMEN (Mitra, Noy, & Jaiswal, 2005) uses a Bayesian Network to
improve the results of the alignment process by deriving unde-
tected correspondences and rejecting existing false positives. By
using probabilistic methods it enhances existing ontology map-
pings by deriving missed matches and invalidating existing
false matches. This system was described in just one paper in
2005 (Mitra et al., 2005).
� OntoDNA (Kiu & Lee, 2007) is an automated and scalable system
that uses hybrid unsupervised clustering techniques, which
include Formal Concept Analysis (FCA) (Formica, 2006), Self-
Organizing Map (SOM) and K-Means clustering, in addition to
a Levenshtein edit distance as lexical measurement. OntoDNA
was first described in 2006 (Kiu & Lee, 2006) and then revisited
in 2007 to present its results in the OAEI (Kiu & Lee, 2007).
� ontoMATCH (Lu, 2010) exploits both semantic and structural
information to compute the alignments. It is internally divided
into five components, a preprocessor, three individual matchers
(Element Matcher, Relationship Matcher and Property
Matcher), a combiner and a selector. This system was described
in one paper in 2010 (Lu, 2010).
� OPTIMA (Thayasivam et al., 2012) uses lexical information from
the concepts to generate a seed alignment. Then it iteratively
searches the space of candidate alignments following the tech-
nique of expectation–maximization until convergence. This sys-
tem was presented in 2011 taking part at the OAEI (Thayasivam
& Doshi, 2011). It also took part in the following year’s contest
(Thayasivam et al., 2012).
� OWL-CM (Yaghlane & Laamari, 2007) uses different matchers to
compute the alignments, whose results are then combined. It
also includes some belief functions into the alignment process
to improve the computed results. In this literature review we
only retrieved one article in 2007 (Yaghlane & Laamari, 2007).
� PRIOR+ (Mao & Peng, 2007) is an ontology matching system that
lays its foundations on propagation theory, information retrie-
val and artificial intelligence. It profits the linguistic and struc-
tural information of the ontologies to match and measures the
profile similarity of different elements in a vector space model.
This system took part in the editions of 2006 (Mao & Peng,
2006) and 2007 (Mao & Peng, 2007) of the OAEI where its
results and performance are described.
� QOM (Ehrig & Staab, 2004) is a variation of the NOM algorithm
devoted to improving the efficiency of the system. Some basic
matchers are used whose results are refined by means of a sig-
moid function, to be lately aggregated and sifted to output the
final alignment. This system was developed in the first years
of the interval considered in this literature review in 2 papers
(Ehrig & Staab, 2004; Ehrig & Sure, 2004).
� RiMOM (Wang et al., 2010) uses three different matching strat-
egies, name-based, metadata-based and instance-based, whose
results are then filtered and combined. A similarity propagation
procedure is iteratively run until no more candidate mappings
are discovered and the system converges. Within the limits of
this review, RiMOM is the system that accounts for the highest
number of individual publications describing it, 9. These articles
span from 2004 to 2013. In this period, there have been periods
of interruption in the flow of publications, however we believe
they account for the development of a new version of the sys-
tem (Li, Li, Zhang, & Tang, 2006; Li, Zhong, Li, & Tang, 2007;
Li, Tang, Li, & Luo, 2009; Tang, Liang, Li, & Wang, 2004; Tang
et al., 2006; Wang et al., 2010; Zhang, Zhong, Li, & Tang,
2008; Zhang, Zhong, Shi, Li, & Tang, 2009). RiMOM has taken
part in several editions of the OAEI, therefore it has been tested
and evaluated over several years.
� SAMBO (Lambrix, Tan, & Liu, 2008) contains several matchers
that exploit different features of the ontologies, and it is the
user the one who decides to use one or several. If several are
chosen the combination of the results by each of them are com-
puted by means of a weighted sum. Results are then filtered
according to some thresholds and presented to the user as sug-
gested alignments to be confirmed. This system has been main-
tained from 2005 to 2008. In this period, at least 5 different
articles were published describing its performance and its
results in the OAEI (Lambrix & Tan, 2005, 2006; Lambrix et al.,
2008; Tan, Jakoniene, Lambrix, Aberg, & Shahmehri, 2006; Tan
& Lambrix, 2007).
� SEMA (Spiliopoulos, Valarakos, Vouros, & Karkaletsis, 2007)
combines six different matching methods whose running
sequence is pre-established and where each method takes as
input the results of the previous methods. This procedure is
iteratively applied until no new mappings are discovered. The
matchers used a are lexical matcher, a latent features matcher,
a vector space model matcher, an instance based matcher a
structural based matcher and a property based matcher. SEMA
was described in 2 different articles in 2007 (Spiliopoulos
et al., 2007), one of them presenting its results in the OAEI con-
test (Spiliopoulos et al., 2007).
� SERIMI (Araújo, de Vries, & Schwabe, 2011c) is a matching sys-
tem developed for instance matching, which is mainly divided
into two phases, a selection one and a disambiguation one. Dur-
ing these phases information retrieval strategies and string
matching techniques are applied. This system was described
in 4 articles between 2011 and 2012 (Araújo, Hidders,
Schwabe, & de Vries, 2011a, 2011b; Araújo, Tran, DeVries,
Hidders, & Schwabe, 2012). In 2011 it also took part in the OAEI
(Araújo et al., 2011c).
� ServOMap (Ba & Diallo, 2013) is a large scale ontology matching
system that also supports multilingual terminologies. It uses an
Ontology Server (ServO) and takes advantage of Information
Retrieval techniques to compute the similarity between the
entities in the input ontologies. This system was recently devel-
oped, in 2012 (Ba & Diallo, 2012a, 2013; Diallo & Ba, 2012).
However it has already taken part in the OAEI (Ba & Diallo,
2012b). This points out that developers are actively working
on the maintenance of this system, therefore new improve-
ments are to be expected, in addition to those already released
in 2013 (Diallo & Kammoun, 2013; Kammoun & Diallo, 2013).
� SIGMa (Lacoste-Julien et al., 2013) is a knowledge base iterative
propagation alignment algorithm that uses both the structural
information from the relationship graph as well as some simi-
larity measures between entity properties. SIGMa is also one
960 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
of the systems that can be considered as young since the first
publication where it is described is from 2012 (Lacoste-Julien
et al., 2012). This system is being actively maintained, and a
new publication was released in 2013 (Lacoste-Julien et al.,
2013).
� S-Match (Giunchiglia, Autayeu, & Pane, 2012) is an open source
semantic matching framework that transforms input tree-like
structures such as catalogs, conceptual models, etc., into light-
weight ontologies to then determine the semantic correspon-
dences between them. It contains the implementation of
several semantic matching algorithms, each one suitable for dif-
ferent purposes. The amount of publications describing S-Match
is one of the highest among those considered for this review, 7.
This system has been maintained since 2003, however it was
not a steady process, as the last publications have intervals
between them of at least 2 years (Giunchiglia, Shvaiko, &
Yatskevich, 2004, 2005a, 2005b; Giunchiglia, Yatskevich, &
Shvaiko, 2007; Giunchiglia et al., 2012; Shvaiko, Giunchiglia, &
Yatskevich, 2009).
� SOBOM (Xu, Wang, Cheng, & Zang, 2010a) is an ontology match-
ing algorithm that uses as start point a set of anchors provided
by a lexical matcher. It also uses the Semantic Inductive Similar-
ity Flooding algorithm to compute the similarity between the
concepts of the sub-ontologies obtained from the anchors.
SOBOM has been described in publications ranging from 2008
to 2012 (Xu, Tao, Zang, & Wang, 2008; Xu et al., 2010a; Xu,
Wang, Cheng, & Zang, 2010b; Xu, Wang, & Liu, 2012), including
those of its participation in several editions of OEAI. In the two-
year period from 2010 to 2012, no publications were retrieved
regarding this system. As the last article dates from 2012, new
updates and enhancements are still to be expected.
� SODA (Zghal, Yahia, Nguifo, & Slimani, 2007) uses linguistic and
structural similarity measures to compute the alignment
between two OWL-DL ontologies which are firstly transformed
into graphs. This graphs undergo two successive phases, first a
linguistic similarity comparison and then a structural similarity
comparison, after which the semantic similarity of the graphs is
obtained. This algorithm outputs the correspondences between
the entities together with their similarity measure values. So
far, only one article describing SODA was found (Zghal et al.,
2007), published in 2007.
� TaxoMap (Hamdi, Safar, Niraula, & Reynaud, 2010) provides an
alignment for two OWL ontologies by exploiting the informa-
tion in the labels of the concepts and the subsumption links that
connect those concepts in the hierarchy. TaxoMap has been
maintained since it was released in 2007 up to 2010. In this
interval, 6 articles were published describing the system and
its participation in the OAEI (Hamdi, Zargayouna, Safar, &
Reynaud, 2008; Hamdi, Safar, Niraula, & Reynaud, 2009;
Hamdi et al., 2010; Safar & Reynaud, 2009; Safar, 2007;
Zargayouna, Safar, & Reynaud, 2007).
� TOAST (Jachnik, Szwabe, Misiorek, & Walkowiak, 2012) is an
ontology matching system based on statistical relational learn-
ing. This system needs a train set from which to learn the
semantics equivalence relation on the basis of partial matches.
TOAST is another system, considered young, as all the
publications describing it were found for 2012 (Szwabe,
Misiorek, & Walkowiak, 2012). However, this system has
already taken part in the OAEI (Jachnik et al., 2012). If the devel-
opers continue with this trend, new articles on this system are
to be expected.
� WeSeE-Match (Paulheim, 2012) exploits the idea of using infor-
mation available on the web to match the ontologies which
would supposedly be the procedure followed by a human trying
to manually match some terms without being an expert in the
domain of the matched terms. Therefore this approach uses a
web search engine to retrieve documents relevant to the con-
cepts to match and compare the results obtained, the more sim-
ilar the search results, the higher the concepts’ similarity value.
This system accounts for few publications, however these
report the results obtained by the system in the OAEI of 2012
and 2013 (Paulheim, 2012; Paulheim & Hertling, 2013).
� WikiMatch (Hertling & Paulheim, 2012a) exploits the use of
Wikipedia’s search engine to obtain documents related to the
concepts being matched. Since there is no duplicity in the titles
names of the articles in Wikipedia for the same language, the
algorithm compares the sets of retrieved titles to obtain the sim-
ilarity between the two concepts. As happens with WeSeE-
Match, WikiMatch has been only described so far in two articles,
although these are quite recent. One of them presenting its over-
all behavior (Hertling & Paulheim, 2012a) and the other report-
ing its results for the OAEI’12 (Hertling & Paulheim, 2012b).
� X-SOM (Curino, Orsi, & Tanca, 2007b) combines the similarity
maps output by different matching algorithms by means of a
neural network and uses logical reasoning and heuristics to
enhance the quality of the mappings. This system was devel-
oped between 2007 and 2010, adding up a total of 4 articles
(Curino, Orsi, & Tanca, 2007a; Curino et al., 2007b; Merlin,
Sorjamaa, Maillet, & Lendasse, 2009; Merlin, Sorjamaa,
Maillet, & Lendasse, 2010), including those that describe its par-
ticipation in OAEI’07.
� YAM++ (Ngo & Bellahsene, 2012a) uses machine learning tech-
niques to discover the mappings between entities in two ontol-
ogies, even if these are not expressed in the same natural
language. It uses matchers at element and structural level. At
element level the similarity is computed by some terminologi-
cal metrics which can be combined by machine learning based
combination methods. At structural level the ontologies are
transformed into graphs and considering the results of the ter-
minological metrics as the starting points, a similarity flooding
algorithm propagation is run. This system has a high level of
maturity as it has been continuously evolving since 2009 up
to 2013 (Duchateau, Coletta, Bellahsene, & Miller, 2009b,
2009a; Ngo, Bellahsene, & Coletta, 2011; Ngo & Bellahsene,
2012a, 2012b, 2013). In this period, at least 6 articles describing
its behavior and overall results in the different editions of the
OAEI have been published. Hence its validity and maturity is
well proven.
Further considering the evolution and maturity degree of these
systems, in Table 5 the amount of articles published regarding each
one of the presented systems is shown. These values show a vary-
ing level of development in the different systems, as the amount of
publications ranges from 1 to 9. In Fig. 8 these results are disaggre-
gated by year. As it suggests, systems are devoted on average 2:6
years of work, which are usually consecutive. However, as hap-
pens, for instance, with ASCO the work was interrupted and
resumed 3 years later. In other systems, as RiMOM, this discontin-
uation in the amount of publications accounts for the development
of a new version.
4.4. ’Processing Frameworks’ category
This category covers two types of publications, articles devoted
to researching the processing and exploiting of the ontology align-
ments (25.85%) and also those that describe some enhanced align-
ment frameworks and alignment formats (74.10%).
Among those articles devoted to processing and exploiting the
alignments, the most common topics were related to (i) ontology
merging (Kim, Kim, & Chung, 2011), i.e., integrating two ontologies
from different sources into a single new one with the information
from both of them, (ii) ontology transformation (Šváb-Zamazal,
Table 5
Amount of articles yearly devoted to each system.
System Articles
AgreementMaker 9
Anchor-Flood 3
AOAS 1
AROMA 8
ASCO 2
ASE 1
ASMOV 5
AUTOMSv2 2
CBW 2
CIDER 3
CODI 2
COMA 5
DSSim 4
Eff2Match 1
FalconAO 5
FBEM 2
FuzzyAlign 2
GeRoMeSuite 4
GLUE 1
GOMMA 3
HCONE 3
Hertuda 2
HotMatch 2
Hmatch 3
IF-MAP 1
iMatch 2
KOSImap 2
LDOA 1
Lily 4
LogMap 6
MaasMatch 3
MapPSO 5
MapSSS 3
MEDLEY 1
MoTo 2
OACAS 2
OLA 3
oMap 4
OMEN 1
OntoDNA 2
ontoMATCH 1
OPTIMA 2
OWL-CM 1
PRIOR+ 2
QOM 2
RiMOM 9
SAMBO 5
SEMA 2
SERIMI 4
ServOMap 5
SIGMa 2
S-Match 7
SOBOM 4
SODA 1
TaxoMap 6
TOAST 2
WeSeE-Match 2
WikiMatch 2
X-SOM 4
YAM++ 6
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 961
Svátek, & Iannone, 2010), that implies expressing an ontology with
respect to another one, (iii) reasoning (Zhang, Lin, Huang, & Wu,
2011), that involves using the correspondences between the ontol-
ogies as rules for reasoning with them, and (iv) alignment argumen-
tation (Trojahn, Quaresma, & Vieira, 2012), that is a way of
explaining the alignments by providing arguments to support or
dismiss them.
Despite the remarkable variety of topics in the articles from this
sub-category, they represent a small percentage of the processing
frameworks category, which in their majority were focused mainly
on defining and developing alignment frameworks such as (Noy &
Musen, 2003), where the process does not finish with the align-
ment but other actions are also available for the user, like manip-
ulating the alignments or performing some of the procedures
previously mentioned.
4.5. ‘Practical Applications’ category
The articles in this category present articles where ontology
matching has been applied to a real-life problem. Within this cat-
egory we found articles devoted to different subjects such as (i)
semantic web and web services (Di Martino, 2009) where most pub-
lications presented ways of using ontology matching for service
discovery or service composition, (ii) P2P systems (Atencia,
Euzenat, Pirrò, & Rousset, 2011), where ontology matching was
used as a way to reduce the semantical heterogeneity between
the queries the users pose to the system and the documents stored,
therefore improving the accuracy of the returned results. Other
fields worth mentioning are (iii) learning systems (Arch-int &
Arch-int, 2013), that focus the use of ontology matching tech-
niques either on narrowing down the distance between the user’s
and the stored documents or as a way to ease the knowledge share
and reuse among users, and last, (iv) multi-agent systems (Mascardi,
Ancona, Bordini, & Ricci, 2011), where the use of ontology match-
ing has always been related to guaranteeing that the different
agents in a communication process could be actually able to inter-
act and achieve the common goals.
In spite of the growing tendency in the development of practical
applications, in general lines, it does not reach the 30% of the
matching systems implemented each year, as Fig. 9 reflects. This
situation is quite remarkable as it suggests that only a slight part
of the matching systems developed have a practical application
in real-life projects. To clarify this situation we have conducted a
survey among ontology matching practitioners, where we asked
them mainly about the future challenges of the field and its appli-
cation in real-life projects. The description and results of this sur-
vey are further detailed in Section 6.
4.6. ‘Evaluation’ category
Out of the articles for this literature review, 38 are devoted to
evaluating the performance of the matching systems. We can split
these articles into two categories regarding the scope of the arti-
cles. There are 14 articles (36.84%) focused on studying the perfor-
mance measures and on proposing different alternatives to
evaluate the matching systems. We have included such articles
in a category named elementary approaches.
The remaining 24 articles (63.16%) delve into evaluation meth-
ods were different existing platforms, systems or benchmarks to
evaluate the matching systems are explored.
Regarding the (i) elementary approaches several articles explore
alternatives to the well-known information retrieval measures of
precision and recall which are used in this field to evaluate respec-
tively the correctness and completeness of the matching systems.
Examples of such publications are the works by Paulheim, Hertling
and Ritze (Paulheim, Hertling, & Ritze, 2013), Niu, Wang, Wu, Qi
and Yu (Niu, Wang, Wu, Qi, & Yu, 2011) or Euzenat (Euzenat,
2007). Additionally in this category we have also included those
papers that describe a new evaluation method or approach such
as the ones by Ferrara, Nikolov, Noessner and Scharffe (Ferrara,
Nikolov, Noessner, & Scharffe, 2013) and by Tordai, van Ossenbrug-
gen, Schreiber and Wielinga (Tordai, van Ossenbruggen, Schreiber,
& Wielinga, 2011).
These measures and approaches are usually included as part of
the systems proposed to evaluate the matching systems which are
included in the (ii) evaluation methods category. In this category we
have included those papers delving into the different existing
Fig. 8. Evolution of the systems over the years.
962 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
platforms, systems and benchmarks used for evaluation, as well as
the results of evaluating the most widespread systems against
these benchmarks or in these platforms.
Regarding the data sets or benchmarks for evaluation, the most
well-known and used are those developed for the Ontology Align-
ment Evaluation Initiative which has been taken as a reference since
2004 (Euzenat, Stuckenschmidt, & Yatskevich, 2005) and that has
been evolving over the years (Rosoiu, dos Santos, & Euzenat,
2011). This initiative contains various tracks with different data
sets which evaluate several features of the tested systems.
The benchmark test (Euzenat, Ros�oiu, & Trojahn, 2013) is built
around a seed ontology and many variations of it, and its purpose
is to provide a stable and detailed picture of the contesting algo-
rithms. These tests are organized into simple tests, where the objec-
tive is to compare the original ontology with itself, a random one
and a generalization, systematic tests, where the original ontology
is to be compared with others where some modifications have
been included, such as removing names, translating into other
languages, flattening or expanding the hierarchy, etc., and finally,
real-life ontologies. The anatomy track evaluates the matching sys-
tems with the task of matching two large ontologies, the Adult
Mouse Anatomy and part of the NCI Thesaurus which describes
the human anatomy. The conference track contains different ontol-
ogies from the conference organization domain. The interest of this
track lies in the fact that these ontologies have been independently
defined. The MultiFarm track aims at testing the ability of the sys-
tems to deal with multilingualism. The library track is a real-world
task to match two thesaurus, the STW and the TheSoz, both used in
libraries for indexation and retrieval. The interactive track tests the
results obtained by the systems when the user is somehow
involved. The Large BioMed track consists of finding alignments
between the Foundational Model of Anatomy (FMA, 2013),
SNOMED Clinical Terms (SNOMED, 2013), and the National Cancer
Institute Thesaurus (NCI, 2013). Finally, the Instance Matching
track focus its efforts on instance matching systems and
techniques.
Fig. 9. Evolution of matching systems vs. practical applications.
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 963
Besides the benchmarks, in this category we have also included
some actual systems used to evaluate the matching systems such
as Wrigley, García-Castro, and Nixon (2012) and Tyl and Loufek
(2009).
In this section we have presented our classification framework
and further detailed the results of classifying the retrieved articles
with this classification framework. In the following section we ana-
lyze the limitations to this review.
5. Limitations of the Literature Review
A literature review in the field of ontology matching is a very
demanding task firstly due to the amount of background knowl-
edge necessary to properly sort the identified articles, and secondly
by the extent of the subject itself and the number of fields where
the research on ontology matching is used.
The articles studied for this review were retrieved by querying
online databases with different expressions regarding the ontology
matching field. In spite of the high amount of articles retrieved,
over 1600, it is possible that many others have not been recovered
as we are dealing with a very wide field of knowledge.
Other databases besides those queried for this review, could
have also been used to raise the amount of retrieved articles and
broaden the scope of the review, however, the databases used
are considered as the most relevant ones among the practitioners.
In addition only those articles in english were finally included even
though some publications were written in other languages, as we
considered english as the predominant language regarding the
research community.
In spite of the limitations previously described, this paper
makes a brief review of the ontology matching field between
2000 and 2013. The articles written in this period were also sorted
according to a classification framework which has allowed us to
identify the different topics and problems that researchers were
tackling for the last decade. Nevertheless this has also brought
up several questions regarding the current research interests of
researchers and practitioners, mainly whether they have continued
to research on the same topics or not, and to check, for instance, if
they have changed topic or even field. Another main issue that we
have detected is the fact that, in the last decade lots of different
matching systems and techniques have been developed, however,
we could not state their use in real-life applications.
In order to clarify these doubts we have designed and per-
formed a survey among the researchers. Its structure as well as
the results of this survey are detailed in Section 6.
6. Trends in Ontology Matching: Practitioner-oriented Survey
We conducted a survey to clarify those concerns emerged from
our literature review. Such concerns were mainly related to the
current state of the research on ontology matching and its applica-
tion in real-life projects.
6.1. Participants and Survey design
The participants in the survey were selected among those tak-
ing part at the OAEI contests. In a two month period, from Decem-
ber 2013 to February 2014, they were individually contacted by
email and presented with the questionnaire shown in Table 6. Even
though the participants were directly contacted by email, their
identities and responses were strictly confidential and only avail-
able to the team conducting the survey. Out of the 288 experts con-
tacted, we received 46 replies.
The survey was designed with 8 short open-ended questions.
Although we have initially considered to define some of these
questions as multiple-choice, we discarded that idea as we did
not want to influence at any degree in the answers provided by
the participants.
These questions can be classified into three groups. Questions 1
to 3 are background questions, questions 4; 5 and 7 are research field
questions and questions 6 and 8 are future challenges questions.
The background questions were included in the questionnaire to
assess the suitability of the participants and contextualize the
answers they may provide. The research field questions were
designed to gain knowledge about the current fields and topics
that have become more attractive to the research community,
and finally future challenges questions were designed to identify
according to practitioners’ point of view the main challenges that
are still to be addressed and the potential expansion fields for
ontology matching.
6.2. Survey results
Out of the 46 answers that we received, only 5 declined the par-
ticipation in the survey, one answered it partially and 13 have not
been researching on ontology matching for a while. Out of this
researchers that have stopped working in the field, some of them
have recently stopped and they answered the questionnaire any-
how, as their contribution was still relevant while others suggested
a more appropriate contact within their groups to redirect the
requests, half of which were answered back. To sum up, the initial
Table 6
Questionnaire used for the survey.
Number Question Type
1. How long have you been researching in Ontology Matching? Background
2. What are your main purposes to do it? Background
3. How many research papers have you written on topics related to Ontology Matching? Background
4. Within the Ontology Matching field, in which particular topic are you currently working on? Research field
5. From your point of view, which are the main fields where the research on Ontology Matching is currently being applied? Research field
6. According to your expertise, which are the main challenges that are still to be addressed? Future challenges
7. Will you continue to research in Ontology Matching?Why? Research field
8. In which fields do you believe that Ontology Matching could also be used? Future challenges
964 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
amount of replies was cut down to 33 actual answers with profit-
able information.
6.2.1. Background questions
The background of the participants in the research is quite
broad and varied, and it includes different types of researchers in
the field. From a temporal point of view, there are those who have
started in the field more than a decade ago but also those who have
recently started to work in this field, specifically values range from
1 to 14 years. Most of the practitioners (78.12%) who answered this
questionnaire have been researching in the field for over 5 years,
being the average number of years working in the field is slightly
superior to 7.
In Fig. 10, the number of years researching in the field is shown
in relation to the number of researchers.
Moreover some researchers have directly tackled the ontology
matching problem by focussing on very specific topics such as
assessing the impact of using different similarity metrics in different
ontology matching tasks, aligning large ontologies or improving ontol-
ogy matching by using reasoning, while others arrived to ontology
matching as a support tool for matters such as data integration,
semantic interoperability or telecommunications systems
interoperability.
Anyhow the suitability of the participants is more than satisfac-
tory as they account on their own for above 460 publications of dif-
ferent types linked to this subject.
6.2.2. Research field questions
The research field questions were included in the questionnaire
to learn about the fields where respondents are working, as well as,
the fields where they think ontology matching could also be
applied. Out of the answers sent by the participants, we could
determine that a high percentage of them (63%) is working in
any of these four topics: instance matching, user involvement in
the matching process, data interlinking and discovery of different
types of correspondences, not 1:1 equivalence relations. The rest of
the respondents mentioned other topics such as parallel ontology
matching, large-scale ontology matching, ontology matching negotia-
tion and mapping reuse, which are more specific. However, besides
these main research topic, most respondents included similarity
metrics and combinations of methods to improve the coverage of
ontology matching as a way to enhance or support their other
research interests.
Regarding the question about the fields where ontology match-
ing is being applied, the consensus shown was noticeable. This
question provided mainly two types of answers. From a practical
point of view, respondents agree that the medical and life science
domain is the one that is using ontology matching the most. Other
researchers offered a more theoretical type of answers, mostly
mentioning data integration and interoperability as the fields where
ontology matching is being applied.
We have found really meaningful that several researchers
pointed out that nowadays the use of ontology matching tech-
niques is reserved to spot cases and that the research at this time
seems merely foundational. However, they also agree that ontol-
ogy matching can be applied in any field where there are two par-
ties that need to communicate and that employ potentially
different protocols, being this way the list of use cases potentially
long.
Finally, when questioned about whether they would continue
to research on this field, the majority, 63.64%, confirmed they
would follow with present or related research lines claiming as
reasons for instance, that there are still plenty of challenges to
address and that the development of new domains will sparkle new
matching problems. On the contrary, 30.30% of the respondents sta-
ted that they would, if not yet, change subjects. Among the reasons
to do so, some mentioned they have moved to other related fields
such as linking open information systems or knowledge transforma-
tion, while others definitely quit the field claiming the lack of use-
fulness for real applications or the little incentive coming from the
application side. A small percentage, 6.06% were still considering
whether to change subject or not.
6.2.3. Future challenges questions
These group of questions were included in the questionnaire to
gain knowledge about how practitioners see the future evolution of
the field and the main challenges still to be addressed. These
answers provided are really useful as they identify a variety of
challenges to address and quote several new fields where matching
techniques could also be used and hence they could be used to
guide the research lines adopted by different research groups.
Regarding the main challenges still to be addressed in the ontol-
ogy matching field, most respondents agree on the need to auto-
matically discover complex relations, instead of 1:1, to correctly
align large ontologies and to focus on applying automatically created
mappings to practical applications.
Other topics that arose in the responses were not supported by
so many respondents but they point out anyway challenges that
need to be addressed, such as:
1. Automated acquisition of reference alignment for evaluating
large scale matching systems.
2. Creating large datasets to asses matching algorithms.
3. Define good tools that are easy to use for non-experts.
4. Develop high quality and fast intelligent combinations of
string-based and new semantic-similarity measures.
5. Holistic ontology matching.
6. How to effectively complement automatic computation with
human validation.
7. How to minimize involvement of users when turning
matches into mapping.
8. Human readable explanations for matches.
9. Improving the mapping process through semi-automatic
machine learning.
10. Integration of domain knowledge into alignment techniques.
11. Learning what metrics to choose in which scenario.
12. Precision and Recall of automatic methods.
13. Scalability and parallelization of the matching.
Fig. 10. Number of researchers in relation with the number of years working in the field.
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 965
14. Semantic mapping.
The answers provided by the practitioners for this question
point out challenges that are inline with those highlighted by Euz-
enat and Shvaiko in Shvaiko and Euzenat (2013) and Euzenat and
Shvaiko (2013).
Table 7 presents a comparison of those challenges outlined by
Euzenat and Shvaiko and those mentioned by respondents to our
survey. As this table states, most of the challenges issued by prac-
titioners have been also considered in those by Euzenat and
Shvaiko, however there are certain mismatches worth noticing.
As previously stated, most researchers mainly agree on 3 chal-
lenges, identified in Table 7 with ‘⁄’, however, just one of these
can be classified into the categories by Euzenat and Shvaiko. It is
worth noticing that one of this challenges is the application of
automatically created mappings to practical applications, which
supports our working hypothesis to start this survey, that the
application of ontology matching techniques in real-life projects
is still a field to be further developed.
Some practitioners, in addition to pointing out the challenges,
also used the answer to this question to mention certain situations
that they consider a mistake, as the fact that most approaches to
ontology matching focus on lexicographic and structural informa-
tion while language is more complex than that and hence achiev-
ing a perfect precision and recall is impossible in real life
applications. Other aspect they complained about was benchmarks
habitually used to test the matching systems (OAEI benchmarks).
They claim that even if they are really useful, their main drawback
is that there are yet too many artificial datasets and tasks in it.
Other practitioners took their remarks a step forward claiming that
science culture does not reward creating and maintaining one tool
and instead everyone creates a prototype for a paper and then
abandons it.
Finally, regarding the fields where ontology matching could also
be applied we have obtain two types of answers. Some practitio-
ners gave a fuzzy answer mentioning that matching techniques
could be used practically anywhere where is no standard for informa-
tion exchange and where the domains are open to adopt ontology
approaches or in broad sense in any information related field.
On the contrary, others actually mentioned fields to apply these
techniques, such as: bioinformatics, information systems, e-com-
merce, web services, intrusion detection systems, cultural heritage,
library science, government, education, banking, personal and social
data management, law, etc. Most agree that the fact that in these
fields ontologies and ontology matching techniques are not already
in use is due to a lack of information regarding the potential
benefits.
7. Limitations of the Survey
There are, of course, limitations to this survey, the foremost
being the sampling size and the population. Although we feel that
our 33 final responses offered a wide variety of useful remarks and
points of view, it is true that the sample is still quite small and
hence our analysis may be biased. Besides, in an effort to prevent
the questions from influencing the answers and to obtain as much
information as possible, we have defined the questionnaire with 8
open-ended questions. This fact, possibly together with the way
some questions were posed, led us to obtaining answers that,
Table 7
Comparison of future challenges.
Challenges identified by Euzenat & Shvaiko Challenges mentioned by practitioners
LARGE-SCALE AND EFFICIENT MATCHING Automated acquisition of reference alignment for evaluating large scale matching systems
Creating large datasets to asses matching algorithms
Scalability and parallelization of the matching
⁄ Correctly align large ontologies
MATCHING WITH BACKGROUND KNOWLEDGE Integration of domain knowledge into alignment techniques
MATCHER SELECTION, COMBINATION AND TUNING Develop high quality and fast intelligent combinations of string-based and new semantic-similarity measures
Learning what metrics to choose in which scenario
USER INVOLVEMENT How to effectively complement automatic computation with human validation
How to minimize involvement of users when turning matches into mapping
EXPLANATION OF MATCHING RESULTS Human readable explanations for matches
UNCERTAINTY IN ONTOLOGY MATCHING –
ALIGNMENT MANAGEMENT –
Other challenges that do not fit in previous categories
Improving the mapping process through semi-automatic machine learning
Precision and Recall of automatic methods
Semantic mapping
Define good tools that are easy to use for non-experts
Holistic ontology matching
⁄ Automatically discover complex relations, instead of 1:1
⁄ Focus on applying automatically created mappings to practical applications
966 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
however interesting, did not exactly match what we expected from
them. Also, the participants targeted were obtained from the par-
ticipants at the OAEI contests, most of which are academically ori-
ented, therefore our survey may be biased towards academical
researchers rather than a balance between academical and indus-
trial researchers. Finally, this survey was the answer to some con-
cerns that arose while conducting the literature review, and we
consider the results here as a first analysis. Our intention is to revi-
sit these answers looking for deeper connections between the
answers, to address questions such as: Is there any relation between
how long a researcher has been working in a subject and the amount
of publications?, Which type of researchers are more prone to
quitting?, etc.
8. Conclusions
In this paper, we have achieved a twofold goal. Initially we per-
formed a literature review of the ontology matching field, whose
results led to the definition and development of the survey lately
conducted.
To address the task of performing a literature review of the
ontology matching field, we have defined a classification frame-
work which helped in structuring our review by providing a com-
prehensive model to sort the different types of publications. This
review was based on an online search of ontology matching related
papers from 2003 to the first semester of 2013. The initial amount
of articles obtained, over 1600, was reduced by filtering them
according to their topics, keywords, abstracts and content.
With the articles left after the several trimming iterations, we
have initially performed a statistical evaluation and analysis. Later,
we have sorted the articles following the framework and then we
have analyzed each one of the categories in the framework, evalu-
ating the different types of articles and topics treated.
While performing this deeper analysis of the articles and their
topics, some concerns arose regarding the actual research interests
of the practitioners as we detected a high amount of papers related
to theoretical solutions and approaches while the number of
applied ones was significantly lower. The approach chosen to clar-
ify these concerns was to ask openly to the research community, by
means of a practitioner-oriented survey.
The purpose of such survey was to gain knowledge about the
current state of the ontology matching field and the application
of such techniques to real-life environments. We have noticed that
most researchers share the same concerns about the practical
application of the ontology matching techniques, and the problem
of having too many theoretical solutions but few applied ones.
However, due to the nature of the survey, with open-ended ques-
tions, there is more information that we have not reflected in this
work and which we plan on analyzing and exploiting in the future.
By means of this work we have provided a general overview of
the ontology matching field in the last decade. It can be used as a
starting point for new practitioners to get a general idea but also, to
help in deciding on research lines, hopefully by tackling some of
the challenges highlighted in the survey.
References
Acampora, G., Loia, V., Salerno, S., & Vitiello, A. (2012). A hybrid evolutionary
approach for solving the ontology alignment problem. International Journal of
Intelligent Systems, 27, 189–216.
ACM. (2013). ACM Digital Library. URL: .
Akbari, I., Fathian, M., & Badie, K. (2009). An improved mlma+ and its application in
ontology matching. In Innovative technologies in intelligent systems and industrial
applications, 2009. CITISIA 2009 (pp. 56–60).
Albagli, S., Ben-Eliyahu-Zohary, R., & Shimony, S. E. (2009). Markov network based
ontology matching. In IJCAI (pp. 1884–1889).
Albagli, S., Ben-Eliyahu-Zohary, R., & Shimony, S. E. (2012). Markov network based
ontology matching. Journal of Computer and System Sciences, 78, 105–118.
Aleksovski, Z., Ten Kate, W., & Van Harmelen, F. (2008). Using multiple ontologies as
background knowledge in ontology matching. In Proceedings of the 1st
international workshop on collective semantics: collective intelligence and the
semantic web CISWeb’08 (pp. 35–49).
Araújo, S., Hidders, J., Schwabe, D., & de Vries, A. P. (2011a). Serimi – resource
description similarity, rdf instance matching and interlinking. CoRR abs/
1107.1104. URL: .
Araújo, S., Hidders, J., Schwabe, D., & de Vries, A. P. (2011b). Serimi – resource
description similarity, rdf instance matching and interlinking. In P. Shvaiko, J.
Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. URL:
.
Araújo, S., Tran, D., DeVries, A., Hidders, J., & Schwabe, D. (2012). Serimi: Class-based
disambiguation for effective instance matching over heterogeneous web data.
In Z. G. Ives, & Y. Velegrakis (Eds.), WebDB (pp. 25–30). URL: .
Araújo, S., de Vries, A.P., & Schwabe, D., (2011c). SERIMI results for OAEI 2011. In P.
Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-
WS.org. (pp. 212–219).
Arch-int, N., & Arch-int, S. (2013). Semantic ontology mapping for interoperability
of learning resource systems using a rule-based reasoning approach. Expert
Systems with Applications, 40, 7428–7443.
Atencia, M., Euzenat, J., Pirrò, G., & Rousset, M. C. (2011). Alignment-based trust for
resource finding in semantic p2p networks. In Proceedings of the 10th
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0005
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0005
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0005
http://dl.acm.org/
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0025
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0025
http://dblp.uni-trier.de/db/journals/corr/corr1107.html#abs-1107-1104
http://dblp.uni-trier.de/db/journals/corr/corr1107.html#abs-1107-1104
http://dblp.uni-trier.de/db/conf/semweb/om2011.html#AraujoHSV11
http://dblp.uni-trier.de/db/conf/webdb/webdb2012.html#AraujoTDHS12
http://dblp.uni-trier.de/db/conf/webdb/webdb2012.html#AraujoTDHS12
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0055
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0055
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0055
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0060
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0060
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 967
international conference on the semantic web – Volume Part I (pp. 51–66). Berlin,
Heidelberg: Springer-Verlag.
Aumüller, D., Do, H. H., Massmann, S., & Rahm, E. (2005). Schema and ontology matching
with coma++. In F. Özcan (Ed.), SIGMOD conference (pp. 906–908). ACM. URL:.
Ba, M., & Diallo, G. (2012a). Knowledge reposiory as entity similarity computing
enabler. In SITIS (pp. 975–981).
Ba, M., & Diallo, G. (2012b). Servomap and servomap-lt results for oaei 2012. In P.
Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt
(Eds.), OM, CEUR-WS.org. URL: .
Ba, M., & Diallo, G. (2013). Large-scale biomedical ontology matching with
ServOMap. IRBM, 34, 56–59.
Behkamal, B., Naghibzadeh, M., & Moghadam, R. (2010). Using pattern detection
techniques and refactoring to improve the performance of ASMOV. In 5th
International symposium on telecommunications (IST) (pp. 979–984).
Bock, J. (2010). Mappso results for oaei 2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia,
H. Stuckenschmidt, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. URL: .
Bock, J., Dänschel, C., & Stumpp, M. (2011). MapPSO and MapEVO results for OAEI
2011. In P. Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM,
CEUR-WS.org. (pp. 179–183).
Bock, J., & Hettenhausen, J. (2008). Mappso results for oaei 2008. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, & H. Stuckenschmidt (Eds.), OM, CEUR-WS.org. URL:
.
Bock, J., & Hettenhausen, J. (2010). Discrete particle swarm optimisation for
ontology alignment. Information Sciences [in Press]. doi:http://dx.doi.org/
10.1016/j.ins.2010.08.013.
Bock, J., Lenk, A., & Dänschel, C., 2010. Ontology Alignment in the cloud. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao, & I. Cruz (Eds.),
Proceedings of the 5th international workshop on ontology matching (OM-2010),
CEUR workshop proceedings. (pp. 73–84).
Bock, J., Liu, P., & Hettenhausen, J. (2009). Mappso results for oaei 2009. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal
(Eds.), OM, CEUR-WS.org. URL: .
Calı̀, A., Lukasiewicz, T., Predoiu, L., & Stuckenschmidt, H. (2008). Tightly integrated
probabilistic description logic programs for representing ontology mappings. In
Proceedings of the 5th international conference on Foundations of information and
knowledge systems (pp. 178–198). Berlin, Heidelberg: Springer-Verlag.
Castano, S., Ferrara, A., Lorusso, D., & Montanelli, S. (2007). The hmatch 2.0 suite for
ontology matchmaking. In G. Semeraro, E. D. Sciascio, C. Morbidoni, & H.
Stoermer (Eds.), SWAP, CEUR-WS.org. URL: .
Castano, S., Ferrara, A., Lorusso, D., & Montanelli, S. (2008). On the ontology instance
matching problem. In 19th International workshop on database and expert
systems application, 2008. DEXA’08 (pp. 180–184).
Castano, S., Ferrara, A., & Messa, G. (2006). Results of the HMatch ontology
matchmaker in OAEI 2006. In P. Shvaiko, J. Euzenat, N. F. Noy, H.
Stuckenschmidt, V. R. Benjamins, & M. Uschold (Eds.), Ontology matching,
CEUR-WS.org. (pp. 134–143).
Cheatham, M. (2011). MapSSS results for OAEI 2011. In P. Shvaiko, J. Euzenat, T.
Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. (pp. 171–179).
Cheatham, M., & Hitzler, P. (2013). Stringsauto and mapsss results for oaei 2013. In
OM (pp. 146–152).
Chua, W. W. K., & Kim, J. J. (2010). Eff2Match results for OAEI 2010. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao, & I.F. Cruz (Eds.), OM,
CEUR-WS.org. (pp. 150–157).
Cohen, W. W., Ravikumar, P. D., & Fienberg, S. E. (2003). A Comparison of string
distance metrics for name-matching tasks. In IIWeb (pp. 73–78).
Cross, V., Stroe, C., Hu, X., Silwal, P., Panahiazar, M., Cruz, I. F., Parikh, P., & Sheth, A.
P. (2011). Aligning the parasite experiment ontology and the ontology for
biomedical investigations using agreementmaker. In O. Bodenreider, M. E.
Martone, & A. Ruttenberg (Eds.), ICBO, CEUR-WS.org. URL: .
Cruz, I. F., Antonelli, F. P., & Stroe, C. (2009a). AgreementMaker: Efficient matching
for large real-world schemas and ontologies. In Proc. VLDB Endow. 2 (pp. 1586–
1589).
Cruz, I. F., Antonelli, F. P., & Stroe, C. (2009b). Agreementmaker: Efficient matching
for large real-world schemas and ontologies. PVLDB 2 (pp. 1586–1589). URL:
.
Cruz, I. F., Antonelli, F. P., & Stroe, C. (2009c). Efficient selection of mappings and
automatic quality-driven combination of matching methods. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. Noy, & A. Rosenthal (Eds.),
Proceedings of the 4th international workshop on ontology matching, CEUR
workshop proceedings (pp. 49–60). URL: .
Cruz, I. F., Antonelli, F. P., Stroe, C., Keles, U. C., & Maduko, A. (2008). Using
agreementmaker to align ontologies for oaei 2009: Overview, results, and
outlook. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, &
A. Rosenthal (Eds.), OM, CEUR-WS.org. URL: .
Cruz, I. F., Stroe, C., Caci, M., Caimi, F., Palmonari, M., Antonelli, F. P., & Keles, U. C.
(2010). Using agreementmaker to align ontologies for oaei 2010. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao, & I. F. Cruz (Eds.), OM,
CEUR-WS.org. URL: .
Cruz, I. F., Stroe, C., Caimi, F., Fabiani, A., Pesquita, C., Couto, F. M., & Palmonari, M.
(2011). Using agreementmaker to align ontologies for oaei 2011. In P. Shvaiko, J.
Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. URL:
.
Cruz, I. F., Stroe, C., Pesquita, C., Couto, F.M., & Cross, V. (2011). Biomedical ontology
matching using the agreementmaker system. In O. Bodenreider, M. E. Martone,
& A. Ruttenberg (Eds.), ICBO, CEUR-WS.org. URL: .
Curino, C., Orsi, G., & Tanca, L. (2007a). X-som: A flexible ontology mapper, In Proc.
of SWAE (DEXA Workshops) (pp. 424–428).
Curino, C., Orsi, G., & Tanca, L. (2007b). X-SOM Results for OAEI 2007. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org. (pp. 276–285).
Dang, T. T., Gabriel, A., Hertling, S., Roskosch, P., Wlotzka, M., Zilke, J. R., Janssen, F., &
Paulheim, H. (2012a). HotMatch results for OEAI 2012. In P. Shvaiko, J. Euzenat,
A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. (pp. 145–151).
Dang, T. T., Gabriel, A., Hertling, S., Roskosch, P., Wlotzka, M., Zilke, J. R., Janssen, F., &
Paulheim, H. (2012b). Hotmatch results for oeai 2012. In P. Shvaiko, J. Euzenat,
A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. URL:.
Dargham, J., & Fares, E. (2008). A hybrid approach for ontology mapping. In
Proceedings of the 2008 international conference on semantic web and web
services, SWWS 2008 (pp. 343–349).
David, J. (2007). AROMA: A method for the discovery of alignments between
ontologies from association rules. Thèse d’informatique. Université de Nantes.
Nantes (FR). URL: .
David, J. (2008a). Aroma results for oaei 2008.
David, J. (2008b). Aroma results for oaei 2009. In P. Shvaiko, J. Euzenat, F.
Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal (Eds.), OM, CEUR-
WS.org. URL: .
David, J. (2011). AROMA results for OAEI 2011. In P. Shvaiko, J. Euzenat, T. Heath, C.
Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. (pp. 122–125).
David, J., Euzenat, J., Scharffe, F., & dos Santos, C. T. (2011). The alignment api 4.0.
Semantic Web 2, 3–10.
David, J., Guillet, F., & Briand, H. (2006). Matching directories and owl ontologies
with aroma. In P. Yu, V. Tsotras, E. Fox, & B. Liu (Eds.), Proc. 15th ACM
international conference on information and knowledge management (CIKM),
Arlington (VA US) (pp. 830–831). US: ACM.
Di Martino, B. (2009). Semantic web services discovery based on structural ontology
matching. International Journal of Web and Grid Services, 5, 46–65.
Diallo, G. & Ba, M. (2012). Effective Method for Large Scale Ontology Matching.
Semantic WebApplications and Tools for Life Sciences (SWAT4LS). CEUR
Workshop Proceedings, 952.
Diallo, G., & Kammoun, A. (2013). Towards learning based strategy for improving
the recall of the servomap matching system. In A. Paschke, A. Burger, P. Romano,
M. S. Marshall, & A. Splendiani (Eds.), SWAT4LS, CEUR-WS.org. URL:.
Direct, S. (2013). Science Direct. URL: .
Do, H., & Rahm, E. (2002). COMA – a system for flexible combination of schema
matching approaches. In Proceedings of the 28th VLDB conference, Hong Kong,
China. URL: .
Do, H. H. (2006). Schema matching and mapping-based data integration: Architecture,
approaches and evaluation. Saarbrücken: Vdm Verlag Dr. Müller.
Doan, A., Madhavan, J., Domingos, P., & Halevy, A. (2004). Ontology matching: A
machine learning approach. In Handbook on ontologies in information systems
(pp. 385–403). Springer-Verlag.
Droge, E. (2010). Guidelines on ontology matching. Information-Wissenschaft und
Praxis, 61, 143–147.
Duchateau, F., Coletta, R., Bellahsene, Z., & Miller, R. J. (2009a). (not) yet another
matcher. In CIKM (pp. 1537–1540).
Duchateau, F., Coletta, R., Bellahsene, Z., & Miller, R. J. (2009b). Yam: A schema
matcher factory. In CIKM (pp. 2079–2080).
Ehrig, M., & Staab, S. (2004). QOM – quick ontology mapping. In Proc. 3rd
international semantic web conference (ISWC04) (pp. 683–697). Springer.
Ehrig, M., & Sure, Y. (2004). Ontology mapping – an integrated approach. In ESWS
(pp. 76–91).
Engmann, D., & Maßmann, S. (2007). Instance matching with coma++. In M. Jarke, T.
Seidl, C. Quix, D. Kensche, S. Conrad, & E. Rahm, et al. (Eds.), BTW workshops
(pp. 28–37). Aachen: Verlagshaus Mainz. URL:.
Esposito, F., Fanizzi, N., & d’Amato, C. (2010). Recovering uncertain mappings
through structural validation and aggregation with the moto system. In SAC (pp.
1428–1432).
Euzenat, J. (2004). State of the art on ontology alignment. Knowledge Web, 2.
Euzenat, J. (2007). Semantic precision and recall for ontology alignment evaluation.
In IJCAI international joint conference on artificial intelligence (pp. 348–353).
Euzenat, J., Guégan, P., & Valtchev, P. (2005). Ola in the oaei 2005 alignment contest.
In Integrating Ontologies.
Euzenat, J., Ros�oiu, M. E., & Trojahn, C. (2013). Ontology matching benchmarks:
Generation, stability, and discriminability. Journal of Web Semantics, 21, 30–48.
Euzenat, J., & Shvaiko, P. (2007). Ontology matching. Berlin, New York: Springer.
Euzenat, J., & Shvaiko, P. (2013). Ontology matching (2nd ed.). Heidelberg (DE):
Springer-Verlag. URL:.
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0060
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0060
http://dblp.uni-trier.de/db/conf/sigmod/sigmod2005.html#AumuellerDMR05
http://dblp.uni-trier.de/db/conf/sigmod/sigmod2005.html#AumuellerDMR05
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#BaD12
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#BaD12
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0085
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0085
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#Bock10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#Bock10
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#BockH08
http://dx.doi.org/10.1016/j.ins.2010.08.013
http://dx.doi.org/10.1016/j.ins.2010.08.013
http://ceur-ws.org
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#BockLH08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#BockLH08
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0125
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0125
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0125
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0125
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0125
http://dblp.uni-trier.de/db/conf/swap/swap2007.html#CastanoFLM07
http://dblp.uni-trier.de/db/conf/swap/swap2007.html#CastanoFLM07
http://dblp.uni-trier.de/db/conf/icbo/icbo2011.html#CrossSHSPCPS11
http://dblp.uni-trier.de/db/conf/icbo/icbo2011.html#CrossSHSPCPS11
http://dblp.uni-trier.de/db/journals/pvldb/pvldb2.html#CruzAS09
http://ceur-ws.org
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#CruzASKM08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#CruzASKM08
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#CruzSCCPAK10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#CruzSCCPAK10
http://dblp.uni-trier.de/db/conf/semweb/om2011.html#CruzSCFPCP11
http://dblp.uni-trier.de/db/conf/icbo/icbo2011.html#CruzSPCC11
http://dblp.uni-trier.de/db/conf/icbo/icbo2011.html#CruzSPCC11
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#DangGHRWZJP12
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#DangGHRWZJP12
http://tel.archives-ouvertes.fr/tel-00200040/en/
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#David08a
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#David08a
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0255
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0255
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0255
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0255
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0260
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0260
http://dblp.uni-trier.de/db/conf/swat4ls/swat4ls2013.html#DialloK13
http://dblp.uni-trier.de/db/conf/swat4ls/swat4ls2013.html#DialloK13
http://www.sciencedirect.com/
http://citeseer.nj.nec.com/do02coma.html
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0285
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0285
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0290
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0290
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0290
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0295
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0295
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0310
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0310
http://dblp.uni-trier.de/db/conf/btw/btw2007w.html#EngmannM07
http://dblp.uni-trier.de/db/conf/btw/btw2007w.html#EngmannM07
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0330
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0345
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0345
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0345
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0350
http://book.ontologymatching.org
968 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
Euzenat, J., Stuckenschmidt, H., & Yatskevich, M. (2005). Introduction to the
ontology alignment evaluation 2005. In Proceedings of the K-Cap 2005 workshop
on integrating ontologies, Banff (Canada) (pp. 97–102).
Euzenat, J., & Valtchev, P. (2004). Similarity-based ontology alignment in OWL-Lite.
In Proc. of the 15th european conference on artificial intelligence (ECAI) – Valencia
(ES) (pp. 333–337).
Falconer, S., Noy, N., & Storey, M. A. (2007). Ontology mapping – a user survey. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), Proceedings of the workshop on
ontology matching (OM2007) at ISWC/ASWC2007, Busan, South Korea (pp. 113–
125).
Fanizzi, N., d’Amato, C., & Esposito, F. (2011). Composite ontology matching with
uncertain mappings recovery. ACM SIGAPP Applied Computing Review, 11, 17–29.
Fernández, S., Velasco, J., Marsa-Maestre, I., & Lopez-Carmona, M. (2012).
FuzzyAlign: A fuzzy method for ontology alignment. In: KEOD 2012 –
proceedings of the international conference on knowledge engineering and
ontology development (pp. 98–107).
Fernández, S., Velasco, J. R., & López-Carmona, M. A. (2009). A fuzzy rule-based
system for ontology mapping. In PRIMA (pp. 500–507).
Ferrara, A., Nikolov, A., Noessner, J., & Scharffe, F. (2013). Evaluation of instance
matching tools: The experience of {OAEI}. Web Semantics: Science, Services and
Agents on the World Wide Web 21, 49–60. Special Issue on Evaluation of
Semantic Technologies.
FMA (2013). Foundational Model of Anatomy. URL: .
Formica, A. (2006). Ontology-based concept similarity in formal concept analysis.
Information Sciences, 176, 2624–2641.
Fu, B., Brennan, R., & O’Sullivan, D. (2009). Multilingual ontology mapping:
Challenges and a proposed framework. In Adaptive and emergent behaviour
and complex systems – proceedings of the 23rd convention of the society for the
study of artificial intelligence and simulation of behaviour, AISB 2009 (pp. 33–35).
Fugazza, C., & Vaccari, L. (2011). Coupling human- and machine-driven mapping of
SKOS thesauri. International Journal of Metadata, Semantics and Ontologies, 6,
155–165.
Gal, A., & Shvaiko, P. (2008). Advances in ontology matching. Lecture notes in
computer science (including subseries lecture notes in artificial intelligence and
lecture notes in bioinformatics) LNCS, (Vol. 4891, pp. 176–198).
Giunchiglia, F., Autayeu, A., & Pane, J. (2012). S-Match: An open source framework
for matching lightweight ontologies. Semantic Web, 3, 307–317.
Giunchiglia, F., Shvaiko, P., & Yatskevich, M. (2004). S-match: An algorithm and an
implementation of semantic matching. In Proc. of the first european semantic web
symposium – ESWS (pp. 61–75).
Giunchiglia, F., Shvaiko, P., & Yatskevich, M. (2005a). S-match: An algorithm and an
implementation of semantic matching. In Semantic Interoperability and
Integration.
Giunchiglia, F., Shvaiko, P., & Yatskevich, M. (2005b). Semantic schema matching. In
OTM conferences (1) (pp. 347–365).
Giunchiglia, F., Yatskevich, M., & Shvaiko, P. (2007). Semantic matching: Algorithms
and implementation. Journal on Data Semantics, 9, 1–38.
Glückstad, F. (2010). Terminological ontology and cognitive processes in
translation. In Proceedings of the 24th Pacific Asia conference on language,
information and computation (pp. 629–636).
Gracia, J., & Asooja, K. (2013). Monolingual and cross-lingual ontology matching
with cider-cl: Evaluation report for oaei 2013. In OM (pp. 109–116).
Gracia, J., Bernad, J., & Mena, E. (2011). Ontology matching with cider: Evaluation
report for oaei 2011. In P. Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I.F.
Cruz (Eds.), OM, CEUR-WS.org. (pp. 1–8).
Gracia, J., & Mena, E. (2008). Ontology matching with cider: Evaluation report for
the oaei 2008. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt
(Eds.), OM, CEUR-WS.org. URL:.
Grau, B. C., Dragisic, Z., Eckert, K., Euzenat, J., Ferrara, A., Granada, R., Ivanova, V.,
Jiménez-Ruiz, E., Kempf, A. O., Lambrix, P., Nikolov, A., Paulheim, H., Ritze, D.,
Scharffe, F., Shvaiko, P., Trojahn, C., & Zamazal, O. (2013). Results of the ontology
alignment evaluation initiative 2013. In: ISWC workshop on ontology matching.
Groß, A., Hartung, M., Kirsten, T., & Rahm, E. (2012). Gomma results for oaei 2012. In
P. Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H.
Stuckenschmidt (Eds.), OM, CEUR-WS.org. URL: .
Haeri, S. H., Abolhassani, H., Qazvinian, V., & Hariri, B. B. (2007). Coincidence-based
scoring of mappings in ontology alignment. JACIII, 11, 803–816. URL: .
Hamdi, F., Safar, B., Niraula, N. B., & Reynaud, C. (2009). Taxomap in the oaei 2009
alignment contest. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt,
N. F. Noy, & A. Rosenthal (Eds.), OM, CEUR-WS.org. URL: .
Hamdi, F., Safar, B., Niraula, N. B., & Reynaud, C. (2010). TaxoMap alignment and
refinement modules: results for OAEI 2010. In P. Shvaiko, J. Euzenat, F.
Giunchiglia, H. Stuckenschmidt, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org.
(pp. 212–219).
Hamdi, F., Zargayouna, H., Safar, B., & Reynaud, C. (2008). Taxomap in the oaei 2008
alignment contest. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt
(Eds.), OM, CEUR-WS.org. URL:.
Hanif, M. S., & Aono, M. (2008a). Alignment results of anchor-flood algorithm for
oaei-2008. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt (Eds.),
OM, CEUR-WS.org. URL:.
Hanif, M. S., & Aono, M. (2008b). Anchor-flood: Results for oaei 2009. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal (Eds.), OM,
CEUR-WS.org. URL:.
Hanif, M. S., & Aono, M. (2009). Anchor-flood: Results for OAEI 2009. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal (Eds.), OM,
CEUR-WS.org. (pp. 127–134).
Hartung, M., Kolb, L., Groß, A., & Rahm, E. (2013). Optimizing similarity
computations for ontology matching – experiences from GOMMA. Lecture
notes in computer science (including subseries lecture notes in artificial intelligence
and lecture notes in bioinformatics) LNBI (Vol. 7970, pp. 81–89).
Hassen, W. (2012). MEDLEY results for OAEI 2012. In P. Shvaiko, J. Euzenat, A.
Kementsietsidis, M. Mao, N.F. Noy, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. (pp. 168–172).
He, W., Yang, X., & Huang, D. (2011). A hybrid approach for measuring semantic
similarity between ontologies based on wordnet. Lecture notes in computer
science (including subseries lecture notes in artificial intelligence and lecture notes
in bioinformatics) LNAI (Vol. 7091, pp. 68–78).
Hertling, S. (2012). Hertuda results for OAEI 2012. In P. Shvaiko, J. Euzenat, A.
Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. (pp. 141–144).
Hertling, S., & Paulheim, H. (2012a). WikiMatch – using Wikipedia for ontology
matching. In P. Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H.
Stuckenschmidt (Eds.), OM, CEUR-WS.org. (pp. 220–225).
Hertling, S., & Paulheim, H. (2012b). Wikimatch results for oeai 2012. In P. Shvaiko,
J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.),
OM, CEUR-WS.org. URL:.
Hu, W., Chen, J., Cheng, G., & Qu, Y. (2010). ObjectCoref and Falcon-AO: Results for
OAEI 2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao,
& I. F. Cruz (Eds.), OM, CEUR-WS.org. URL:.
Hu, W., Cheng, G., Zheng, D., Zhong, X., & Qu, Y. (2006). The results of falcon-ao in
the oaei 2006 campaign. In P. Shvaiko, J. Euzenat, N. F. Noy, H. Stuckenschmidt,
V. R. Benjamins, & M. Uschold (Eds.), Ontology matching, CEUR-WS.org. URL:
.
Hu, W., Jian, N., Qu, Y., & Wang, Y. (2005). GMO: A Graph Matching for Ontologies. In
B. Ashpole, M. Ehrig, J. Euzenat, & H. Stuckenschmidt (Eds.), Integrating
ontologies, CEUR-WS.org. (pp. 41–48).
Hu, W., & Qu, Y. (2008). Falcon-AO: A practical ontology matching system. Web
Semantics: Science, Services and Agents on the World Wide Web, 6, 237–239.
Hu, W., Zhao, Y., Li, D., Cheng, G., Wu, H., & Qu, Y. (2007). Falcon-ao: Results for oaei
2007. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org.
URL: .
Huber, J., Sztyler, T., Nößner, J., & Meilicke, C. (2011). CODI: combinatorial
optimization for data integration: Results for OAEI 2011. In P. Shvaiko, J.
Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. (pp.
134–141).
IEEXplore. (2013). IEEEXplore Digital Library. URL: .
Jachnik, A., Szwabe, A., Misiorek, P., & Walkowiak, P. (2012). TOAST results for OAEI
2012. In P. Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N.F. Noy, & H.
Stuckenschmidt (Eds.), OM, CEUR-WS.org. (pp. 205–212).
Jean-Mary, Y. R., & Kabuka, M. R. (2007). Asmov results for oaei 2007. In P. Shvaiko,
J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org. URL: .
Jean-Mary, Y. R., & Kabuka, M. R. (2008). Asmov: Results for oaei 2008. In P. Shvaiko,
J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt (Eds.), OM, CEUR-WS.org. URL:
.
Jean-Mary, Y. R., Shironoshita, E. P., & Kabuka, M. R. (2008). Asmov: Results for oaei
2009. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A.
Rosenthal (Eds.), OM, CEUR-WS.org. URL: .
Jean-Mary, Y. R., Shironoshita, E. P., & Kabuka, M. R. (2009). Ontology matching with
semantic verification. Web Semantics, 7, 235–251.
Jean-Mary, Y. R., Shironoshita, E. P., & Kabuka, M. R. (2010). Asmov: Results for oaei
2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao, & I. F.
Cruz (Eds.), OM, CEUR-WS.org. URL:
Jian, N., Hu, W., Cheng, G., & Qu, Y. (2005). Falconao: Aligning ontologies with
falcon. In B. Ashpole, M. Ehrig, J. Euzenat, & H. Stuckenschmidt (Eds.), Integrating
Ontologies, CEUR-WS.org. URL:.
Jiménez-Ruiz, E., & Cuenca Grau, B. (2011). LogMap: Logic-based and scalable
ontology matching. Lecture notes in computer science (including subseries lecture
notes in artificial intelligence and lecture notes in bioinformatics) LNCS, (Vol. 7031,
pp. 273–288).
Jiménez-Ruiz, E., Grau, B. C., & Horrocks, I. (2012). Logmap and logmaplt results for
oaei 2012. In P. Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H.
Stuckenschmidt (Eds.), OM, CEUR-WS.org. URL: .
Jiménez-Ruiz, E., Grau, B. C., & Zhou, Y. (2011). Logmap 2.0: Towards logic-based,
scalable and interactive ontology matching. In A. Paschke, A. Burger, P.
Romando, M. S. Marshall, & A. Splendiani (Eds.), SWAT4LS (pp. 45–46). ACM.
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0375
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0375
http://sig.biostr.washington.edu/projects/fm/
http://sig.biostr.washington.edu/projects/fm/
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0400
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0400
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0410
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0410
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0410
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0420
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0420
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0440
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0440
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#GraciaM08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#GraciaM08
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#GrossHKR12
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#GrossHKR12
http://dblp.uni-trier.de/db/journals/jaciii/jaciii11.html#HosseinAQH07
http://dblp.uni-trier.de/db/journals/jaciii/jaciii11.html#HosseinAQH07
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#HamdiSNR08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#HamdiSNR08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#HamdiZSR08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#HamdiZSR08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#HanifA08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#HanifA08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#HanifA08a
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#HanifA08a
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#HertlingP12a
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#HertlingP12a
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#HuCCQ10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#HuCCQ10
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#HuCZZQ06
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0555
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0555
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#HuZLCWQ07
http://ieeexplore.ieee.org/Xplore/home.jsp
http://ieeexplore.ieee.org/Xplore/home.jsp
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#Jean-MaryK07
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#Jean-MaryK07
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#Jean-MaryK08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#Jean-MarySK08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#Jean-MarySK08
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0595
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0595
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#Jean-MarySK10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#Jean-MarySK10
http://dblp.uni-trier.de/db/conf/kcap/kcap2005w.html#JianHCQ05
http://dblp.uni-trier.de/db/conf/kcap/kcap2005w.html#JianHCQ05
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#Jimenez-RuizGH12
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#Jimenez-RuizGH12
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 969
URL:..
Jiménez-Ruiz, E., Meilicke, C., Grau, B. C., & Horrocks, I. (2013). Evaluating mapping
repair systems with large biomedical ontologies. In 26th International workshop
on description logics (pp. 246–257).
Jiménez-Ruiz, E., Morant, A., & Grau, B. C. (2011). Logmap results for oaei 2011. In P.
Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-
WS.org. URL: .
Joslyn, C., Paulson, P., & White, A. M. (2009). Measuring the structural preservation
of semantic hierarchy alignment. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H.
Stuckenschmidt, N. F. Noy, & A. Rosenthal (Eds.), OM, CEUR-WS.org. (pp. 1–12).
Kachroudi, M., Moussa, E. B., Zghal, S., & Yahia, S. B. (2011). LDOA results for OAEI
2011. In P. Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM,
CEUR-WS.org. (pp. 148–156).
Kalfoglou, Y., & Schorlemmer, M. (2003a). IF-Map: An ontology-mapping method
based on information-flow theory. Journal of Data Semantics, 98–127.
Kalfoglou, Y., & Schorlemmer, M. (2003b). Ontology mapping: The state of the art.
Science, 2, 3.
Kammoun, A., & Diallo, G. (2013). Servomap results for oaei 2013. In OM (pp. 169–
176).
Kengue, J. F. D., Euzenat, J., & Valtchev, P. (2007). OLA in the OAEI 2007 evaluation
contest. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-
WS.org. (pp. 188–195).
Kensche, D., Quix, C., Li, X., & Li, Y. (2007). Geromesuite: A system for holistic
generic model management. In C. Koch, J. Gehrke, M. N. Garofalakis, D.
Srivastava, K. Aberer, A. Deshpande, D. Florescu, C. Y. Chan, V. Ganti, C. C. Kanne,
W. Klas, & E. J. Neuhold (Eds.), VLDB (pp. 1322–1325). ACM. URL: ..
Kim, J., Kim, P., & Chung, H. (2011). Ontology construction using online ontologies
based on selection, mapping and merging. International Journal of Web and Grid
Services, 7, 170–189.
Kirsten, T., Gross, A., Hartung, M., & Rahm, E. (2011). Gomma: A component-based
infrastructure for managing and analyzing life science ontologies and their
evolution. Journal of Biomedical Semantics, 2, 6.
Kiu, C. C., & Lee, C. S. (2006). Ontology mapping and merging through ontodna for
learning object reusability. Educational Technology and Society, 9, 27–42. URL:
.
Kiu, C. C., & Lee, C. S. (2007). OntoDNA: Ontology alignment results for OAEI 2007. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org. (pp. 196–205).
Kotis, K., Katasonov, A., & Leino, J. (2012a). ASE results for OAEI 2012. In P. Shvaiko, J.
Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM,
CEUR-WS.org. (pp. 116–123).
Kotis, K., Katasonov, A., & Leino, J. (2012b). AUTOMSv2 results for OAEI 2012. In P.
Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt
(Eds.), OM, CEUR-WS.org. (pp. 124–132).
Kotis, K., & Lanzenberger, M. (2008). Ontology matching: Current status, dilemmas
and future challenges. In International conference on complex, intelligent and
software intensive systems, 2008. CISIS 2008 (pp. 924–927).
Kotis, K., Valarakos, A. G., & Vouros, G. A. (2006). Automs: Automated ontology
mapping through synthesis of methods. In P. Shvaiko, J. Euzenat, N. F. Noy, H.
Stuckenschmidt, V. R. Benjamins, & M. Uschold (Eds.), Ontology matching, CEUR-
WS.org.
Kotis, K., & Vouros, G. A. (2004). The HCONE approach to ontology merging. In
Proceedings of the 21st international symposium on computer architecture
(pp. 137–151). IEEE Computer Society Press.
Kotis, K., Vouros, G. A., & Stergiou, K. (2006). Towards automatic merging of domain
ontologies: The hcone-merge approach. Journal of Web Semantics, 4, 60–79. URL:
.
Lacoste-Julien, S., Palla, K., Davies, A., Kasneci, G., Graepel, T., & Ghahramani, Z.
(2012). Sigma: Simple greedy matching for aligning large knowledge bases.
CoRR abs/1207.4525.
Lacoste-Julien, S., Palla, K., Davies, A., Kasneci, G., Graepel, T., & Ghahramani, Z.
(2013). SIGMa: simple greedy matching for aligning large knowledge bases. In
Proceedings of the 19th ACM SIGKDD international conference on Knowledge
discovery and data mining (pp. 572–580). New York, NY, USA: ACM. http://
dx.doi.org/10.1145/2487575.2487592.
Lai, J., Wang, Y., Zhang, R., Gu, X., Yu, T., & Li, J., 2010. Aggregating multiple ontology
similarity based on iowa operator. In 2010 2nd International workshop on
database technology and applications (DBTA) (pp. 1–4).
Lambrix, P., & Tan, H. (2005). A framework for aligning ontologies. In F. Fages & S.
Soliman (Eds.), PPSWR (pp. 17–31). Springer. URL: .
Lambrix, P., & Tan, H. (2006). Sambo – a system for aligning and merging biomedical
ontologies. Journal of Web Semantics, 4, 196–206.
Lambrix, P., Tan, H., & Liu, Q. (2008). SAMBO and SAMBOdtf Results for the Ontology
Alignment Evaluation Initiative 2008. Proceedings of the 3rd International
Workshop on Ontology Matching (OM-2008). CEUR Workshop Proceedings,
431,190–198.
Lauser, B., Johannsen, G., Caracciolo, C., Van Hage, W., Keizer, J., & Mayr, P. (2008).
Comparing human and automatic thesaurus mapping approaches in the
agricultural domain. In Proceedings of the international conference on dublin
core and metadata applications (pp. 43–53).
Le, B. T., Dieng-Kuntz, R., & Gandon, F. (2004). On ontology matching problems – for
building a corporate semantic web in a multi-communities organization. In
ICEIS (4) (pp. 236–243).
Li, J., Tang, J., Li, Y., & Luo, Q. (2009). RiMOM: A dynamic multistrategy ontology
alignment framework. IEEE Transactions on Knowledge and Data Engineering, 21,
1218–1232. doi: http://dx.doi.org/10.1109/TKDE.2008.202.
Li, Y., Li, J., Zhang, D., & Tang, J. (2006). Result of ontology alignment with rimom at
oaei’06. In P. Shvaiko, J. Euzenat, N. F. Noy, H. Stuckenschmidt, V. R. Benjamins,
& M. Uschold (Eds.), Ontology matching, CEUR-WS.org. URL:
Li, Y., Zhong, Q., Li, J., & Tang, J. (2007). Result of ontology alignment with rimom at
oaei’07. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org.
URL: .
Lin, F., & Sandkuhl, K. (2007). Polygon-based similarity aggregation for ontology
matching. Lecture notes in computer science (including subseries lecture notes in
artificial intelligence and lecture notes in bioinformatics) LNCS (Vol. 4743, pp. 255–
264).
Lin, F., & Sandkuhl, K. (2008a). A survey of exploiting WordNet in
ontology matching. IFIP International Federation for Information Processing, 276,
341–350.
Lin, F., & Sandkuhl, K. (2008b). User-based constraint strategy in ontology matching.
Lecture notes in computer science (including subseries lecture notes in artificial
intelligence and lecture notes in bioinformatics) LNCS, (Vol. 5061, pp. 687–696).
Liu, B., Cao, S. G., Cao, D. F., Li, Q. C., Liu, H. T., & Shi, S. N. (2012). An ontology based
semantic heterogeneity measurement framework for optimization in
distributed data mining. In 2012 International conference on machine learning
and cybernetics (ICMLC) (pp. 118–123).
Loia, V., Fenza, G., De Maio, C., & Salerno, S. (2013). Hybrid methodologies to foster
ontology-based knowledge management platform. In 2013 IEEE symposium on
intelligent agent (IA) (pp. 36–43).
Lu, Z. (2010). ontoMATCH: A probabilistic architecture for ontology matching. In
Conference on information sciences and interaction sciences (ICIS), 2010 3rd
international (pp. 174–180).
Mao, M., & Peng, Y. (2006). Prior system: Results for oaei 2006. In Ontology matching.
Mao, M., & Peng, Y. (2007). The PRIOR+: Results for OAEI campaign 2007. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org. (pp. 219–
226).
Mascardi, V., Ancona, D., Bordini, R., & Ricci, A. (2011). Cool-agentspeak: Enhancing
agentspeak-dl agents with plan exchange and ontology services. In International
conference on web intelligence and intelligent agent technology (WI-IAT), 2011
IEEE/WIC/ACM (pp. 109–116).
Mascardi, V., Locoro, A., & Rosso, P. (2010). Automatic ontology matching via upper
ontologies: A systematic evaluation. IEEE Transactions on Knowledge and Data
Engineering, 22, 609–623.
Massmann, S., Engmann, D., & Rahm, E. (2006). Coma++: Results for the ontology
alignment contest oaei 2006. In P. Shvaiko, J. Euzenat, N. F. Noy, H.
Stuckenschmidt, V. R. Benjamins, & M. Uschold (Eds.), Ontology matching,
CEUR-WS.org. URL: .
Maßmann, S., Raunich, S., Aumüller, D., Arnold, P., & Rahm, E. (2011). Evolution of
the COMA match system. In P. Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I.
F. Cruz (Eds.), OM, CEUR-WS.org. (pp. 49–60).
Merlin, P., Sorjamaa, A., Maillet, B., & Lendasse, A. (2009). X-som and l-som: a nested
approach for missing value imputation. In ESANN. URL: .
Merlin, P., Sorjamaa, A., Maillet, B., & Lendasse, A. (2010). X-som and l-som: A
double classification approach for missing value imputation. Neurocomputing,
73, 1103–1108. URL: .
Mitra, P., Noy, N. F., & Jaiswal, A. R. (2005). Ontology mapping discovery with
uncertainty. In Fourth international conference on the semantic web (ISWC-2005)
(pp. 537–547).
Nagy, M., Vargas-Vera, M., & Motta, E. (2006). Dssim-ontology mapping with
uncertainty. In P. Shvaiko, J. Euzenat, N. F. Noy, H. Stuckenschmidt, V. R.
Benjamins, & M. Uschold (Eds.), Ontology matching, CEUR-WS.org. URL:.
Nagy, M., Vargas-Vera, M., & Motta, E. (2007). Dssim – managing uncertainty on the
semantic web. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-
WS.org. URL: .
Nagy, M., Vargas-Vera, M., & Stolarski, P. (2009). DSSim results for OAEI 2009. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal
(Eds.), OM, CEUR-WS.org. (pp. 160–169).
Nagy, M., Vargas-Vera, M., Stolarski, P., & Motta, E. (2008). Dssim results for oaei
2008. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt (Eds.), OM,
CEUR-WS.org. URL: .
Nasir, S., & Noor, N. (2010). Analysing the effectiveness of COMA++ on the mapping
between traditional Malay textile (TMT) knowledge model and CIDOC CRM. In
Proceedings 2010 international symposium on information technology – visual
informatics, ITSim’10 1 (pp. 1–6).
NCI (2013). National Cancer Institute Thesaurus. URL: .
Ngo, D., & Bellahsene, Z. (2012a). YAM++: A multi-strategy based approach for
ontology matching task. Lecture notes in computer science (including subseries
lecture notes in artificial intelligence and lecture notes in bioinformatics) LNAI (Vol.
7603, pp. 421–425).
Ngo, D., & Bellahsene, Z. (2012b). Yam++ results for oaei 2012. In OM.
Ngo, D., & Bellahsene, Z. (2013). Yam++ results for oaei 2013. In OM (pp. 211–218).
Ngo, D., Bellahsene, Z., & Coletta, R. (2011). Yam++ results for oaei 2011. In OM.
http://dblp.uni-trier.de/db/conf/swat4ls/swat4ls2011.html#Jimenez-RuizGZ11
http://dblp.uni-trier.de/db/conf/swat4ls/swat4ls2011.html#Jimenez-RuizGZ11
http://dblp.uni-trier.de/db/conf/semweb/om2011.html#Jimenez-RuizMG11
http://dblp.uni-trier.de/db/conf/semweb/om2011.html#Jimenez-RuizMG11
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0645
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0645
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0650
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0650
http://dblp.uni-trier.de/db/conf/vldb/vldb2007.html#KenscheQLL07
http://dblp.uni-trier.de/db/conf/vldb/vldb2007.html#KenscheQLL07
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0670
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0670
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0670
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0675
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0675
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0675
http://dblp.uni-trier.de/db/journals/ets/ets9.html#KiuL06
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0710
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0710
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0710
http://dblp.uni-trier.de/db/journals/ws/ws4.html#KotisVS06
http://dx.doi.org/10.1145/2487575.2487592
http://dx.doi.org/10.1145/2487575.2487592
http://dblp.uni-trier.de/db/conf/ppswr/ppswr2005.html#LambrixT05
http://dblp.uni-trier.de/db/conf/ppswr/ppswr2005.html#LambrixT05
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0740
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0740
http://dx.doi.org/10.1109/TKDE.2008.202
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#LiLZT06
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#LiLZT06
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#LiZLT07
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0780
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0780
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0780
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0820
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0820
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0820
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#MassmannER06
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#MassmannER06
http://dblp.uni-trier.de/db/conf/esann/esann2009.html#MerlinSML09
http://dblp.uni-trier.de/db/conf/esann/esann2009.html#MerlinSML09
http://dblp.uni-trier.de/db/journals/ijon/ijon73.html#MerlinSML10
http://dblp.uni-trier.de/db/journals/ijon/ijon73.html#MerlinSML10
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#NagyVM06
http://dblp.uni-trier.de/db/conf/semweb/om2006.html#NagyVM06
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#NagyVM07a
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#NagyVM07a
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#NagyVSM08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#NagyVSM08
http://ncit.nci.nih.gov/
970 L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971
Ngo, D., Bellahsene, Z., & Todorov, K. (2013). Opening the black box of ontology
matching. Lecture notes in computer science (including subseries lecture notes in
artificial intelligence and lecture notes in bioinformatics) LNCS (Vol. 7882, pp. 16–
30).
Niu, X., Wang, H., Wu, G., Qi, G., & Yu, Y. (2011). Evaluating the stability and
credibility of ontology matching methods. Lecture notes in computer science
(including subseries lecture notes in artificial intelligence and lecture notes in
bioinformatics) LNCS (Vol. 6643, pp. 275–289).
Noessner, J., & Niepert, M. (2010). Codi: Combinatorial optimization for data
integration: Results for oaei 2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H.
Stuckenschmidt, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. URL:
Noy, N. F., & Musen, M. A. (2003). The {PROMPT} suite: Interactive tools for ontology
merging and mapping. International Journal of Human-Computer Studies, 59,
983–1024.
OAEI (2013). Ontology Alignment Evaluation Initiative. URL: .
Paulheim, H. (2012). WeSeE-Match results for OEAI 2012. In P. Shvaiko, J. Euzenat,
A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. (pp. 213–219).
Paulheim, H., & Hertling, S. (2013). Wesee-match results for oaei 2013. In OM (pp.
197–202).
Paulheim, H., Hertling, S., & Ritze, D. (2013). Towards evaluating interactive
ontology matching tools. Lecture notes in computer science (including subseries
lecture notes in artificial intelligence and lecture notes in bioinformatics) LNCS (Vol.
7882, pp. 31–45).
Pesquita, C., Stroe, C., Cruz, I. F., & Couto, F. M. (2010). Blooms on agreementmaker:
results for oaei 2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt,
M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. URL: .
Qazvinian, V., Abolhassani, H., (Hossein), S. H. H., & Hariri, B. B. (2008). Evolutionary
coincidence-based ontology mapping extraction. Expert Systems, 25,
221–236.
Quix, C., Gal, A., Sagi, T., & Kensche, D. (2010). An integrated matching system
GeRoMeSuite and SMB: results for OAEI 2010. In P. Shvaiko, J. Euzenat, F.
Giunchiglia, H. Stuckenschmidt, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org.
(pp. 166–171).
Quix, C., Geisler, S., Kensche, D., & Li, X. (2008). Results of GeRoMeSuite for OAEI
2008. Proceedings of the 3rd International Workshop on Ontology Matching
(OM-2008). CEUR Workshop Proceedings, 431, 160–166.
Quix, C., Geisler, S., Kensche, D., & Li, X. (2009). Results of geromesuite for oaei 2009.
In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A.
Rosenthal (Eds.), OM, CEUR-WS.org. URL: .
Reul, Q., & Pan, J. (2010). KOSImap: Use of description logic reasoning to align
heterogeneous ontologies. In CEUR workshop proceedings 573 (pp. 489–500).
Reul, Q., & Pan, J. Z. (2009). Kosimap: Ontology alignments results for oaei 2009. In
P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A.
Rosenthal (Eds.), OM, CEUR-WS.org. URL: .
Rosoiu, M. E., dos Santos, C. T., & Euzenat, J. (2011). Ontology matching benchmarks:
generation and evaluation. In P. Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, &
I. F. Cruz (Eds.), OM, CEUR-WS.org. (pp. 1–12).
Rubiolo, M., Caliusco, M., Stegmayer, G., Coronel, M., & Fabrizi, M. G. (2012).
Knowledge discovery through ontology matching: An approach based on an
artificial neural network model. Information Sciences, 194, 107–119. Intelligent
Knowledge-Based Models and Methodologies for Complex Information
Systems.
Safar, B. (2007). Exploiting wordnet as background knowledge. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, & B. He (Eds.), Proceedings of the workshop on ontology
matching (OM2007) at ISWC/ASWC2007, Busan, South Korea.
Safar, B., & Reynaud, C. (2009). Alignement d’ontologies basé sur des ressources
complémentaires illustration sur le système taxomap. Technique et Science
Informatiques, 28, 1211–1232. URL: .
Sánchez-Ruiz, A., Ontañón, S., González-Calero, P., & Plaza, E. (2011). Measuring
similarity in description logics using refinement operators. Lecture notes in
computer science (including subseries lecture notes in artificial intelligence and
lecture notes in bioinformatics) LNAI (Vol. 6880, pp. 289–303).
Schadd, F. C., & Roos, N. (2011). Maasmatch results for oaei 2011. In P. Shvaiko, J.
Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. URL:
.
Schadd, F. C., & Roos, N. (2012a). MaasMatch results for OAEI 2012. In P. Shvaiko, J.
Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM,
CEUR-WS.org. (pp. 160–167).
Schadd, F. C., & Roos, N. (2012b). Maasmatch results for oaei 2012. In P. Shvaiko, J.
Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt (Eds.), OM,
CEUR-WS.org. URL: .
Schadd, F. C., & Roos, N. (2013). Summary of the maasmatch participation in the
oaei-2013 campaign. In OM (pp. 139–145).
Scharffe, F., Zamazal, O., & Fensel, D. (2013). Ontology alignment design patterns.
Knowledge and Information Systems, 1–28.
Scopus (2013). Scopus. URL: .
Shah, G., & Syeda-Mahmood, T. (2004). Searching databases for sematically-related
schemas. In Proceedings of the 27th annual international ACM SIGIR conference on
research and development in information retrieval (pp. 504–505). New York, NY,
USA: ACM.
Shvaiko, P., & Euzenat, J. (2005). A survey of schema-based matching approaches.
Lecture notes in computer science (including subseries lecture notes in artificial
intelligence and lecture notes in bioinformatics) LNCS (Vol. 3730, pp. 146–171).
Cited By (since 1996)243.
Shvaiko, P., & Euzenat, J. (2013). Ontology matching: State of the art and future
challenges. IEEE Transactions on Knowledge and Data Engineering, 25, 158–176.
Shvaiko, P., Giunchiglia, F., & Yatskevich, M. (2009). Semantic matching with s-
match. In R. D. Virgilio, F. Giunchiglia, & L. Tanca (Eds.), Semantic Web
Information Management (pp. 183–202). Springer. URL: .
SNOMED. (2013). SNOMED Clinical Terms. URL: .
Spiliopoulos, V., Valarakos, A.G., Vouros, G. A., & Karkaletsis, V. (2007). SEMA:
Results for the ontology alignment contest OAEI 2007. In P. Shvaiko, J. Euzenat,
F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org. (pp. 244–254).
Spiliopoulos, V., Vouros, G. A., & Karkaletsis, V. (2010). On the discovery of
subsumption relations for the alignment of ontologies. Web Semantics: Science,
Services and Agents on the World Wide Web, 8, 69–88.
Stoermer, H., & Rassadko, N. (2009a). Results of OKKAM feature based entity
matching algorithm for instance matching contest of OAEI 2009. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal (Eds.), OM,
CEUR-WS.org. (pp. 200–207).
Stoermer, H., & Rassadko, N. (2009b). Results of okkam feature based entity
matching algorithm for instance matching contest of oaei 2009. In OM.
Stoermer, H., Rassadko, N., & Vaidya, N. (2010). Feature-based entity matching: The
fbem model, implementation, evaluation. In CAiSE (pp. 180–193).
Straccia, U., & Troncy, R. (2005a). omap: An implemented framework for
automatically aligning owl ontologies. In SWAP.
Straccia, U., & Troncy, R. (2005b). oMAP: Combining classifiers for aligning OWL
ontologies. In Proceedings of the 6th international conference on web information
systems engineering (WISE 2005), New York City, USA. (pp. 133–147). URL:
.
Straccia, U., & Troncy, R. (2005c). oMAP: Results of the ontology alignment contest.
In Proceedings of the K-Cap workshop on integrating ontologies (IntOnt 2005),
Banff, Canada (pp. 92–96).
Straccia, U., & Troncy, R. (2006). Towards distributed information retrieval in the
semantic web: Query reformulation using the omap framework. In ESWC (pp.
378–392).
Sunna, W., & Cruz, I. F. (2007). Using the agreementmaker to align ontologies for the
oaei campaign 2007. In Shvaiko, P., Euzenat, J., Giunchiglia, F., & He, B. (Eds.),
OM, CEUR-WS.org. URL: .
Šváb-Zamazal, O., Svátek, V., & Iannone, L. (2010). Pattern-based ontology
transformation service exploiting oppl and owl-api. Lecture notes in computer
science (including subseries lecture notes in artificial intelligence and lecture notes
in bioinformatics) LNAI (Vol. 6317, pp. 105–119).
Szwabe, A., Misiorek, P., & Walkowiak, P. (2012). Tensor-based relational learning
for ontology matching. In KES (pp. 509–518).
Tan, H., Jakoniene, V., Lambrix, P., Aberg, J., & Shahmehri, N. (2006). Alignment of
biomedical ontologies using life science literature. In E. G. Bremer, J. Hakenberg,
E. H. Han, D. P. Berrar, & W. Dubitzky (Eds.), KDLL (pp. 1–17). Springer. URL:
.
Tan, H., & Lambrix, P. (2007). Sambo results for the ontology alignment evaluation
initiative 2007. In OM.
Tang, J., Li, J., Liang, B., Hunag, X., Li, Y., & Wang, K. (2006). Using Bayesian decision
for ontology mapping. Web Semantics: Science, Services and Agents on the World
Wide Web, 4, 243–262. http://dx.doi.org/10.1016/j.websem.2006.06.001.
Tang, J., Liang, B. Y., Li, J., & Wang, K. (2004). Risk minimization based ontology
mapping. In C. H. Chi & K. Y. Lam (Eds.), Proceedings of the advanced workshop on
content computing (pp. 469–480). Berlin, Heidelberg: Springer. http://dx.doi.org/
10.1007/978-3-540-30483-8_58.
Thanh Le, B., & Dieng-Kuntz, R. (2007). A graph-based algorithm for alignment of owl
ontologies. In Proceedings of the IEEE/WIC/ACM international conference on web
intelligence (pp. 466–469). Washington, DC, USA: IEEE Computer Society. http://
dx.doi.org/10.1109/WI.2007.10. URL:.
Thayasivam, U., Chaudhari, T., & Doshi, P., 2012. Optima+ results for OAEI 2012. In P.
Shvaiko, J. Euzenat, A. Kementsietsidis, M. Mao, N. F. Noy, & H. Stuckenschmidt
(Eds.), OM, CEUR-WS.org. (pp. 181–188).
Thayasivam, U., & Doshi, P. (2011). Optima results for oaei 2011. In OM.
Tian, X., & Guo, Y. (2010). A cosine theorem based algorithm for similarity
aggregation of ontologies. In 2010 2nd International conference on signal
processing systems (ICSPS) (pp. V2-16–V2-19).
Todorov, K., Geibel, P., & Kühnberger, K. U. (2010). Mining concept similarities for
heterogeneous ontologies. Lecture notes in computer science (including subseries
lecture notes in artificial intelligence and lecture notes in bioinformatics) LNAI (Vol.
6171, pp. 86–100).
Tomaszewski, B., & Holden, E. (2012). The geographic information science and
technology and information technology bodies of knowledge: An ontological
alignment. In Proceedings of the 13th annual conference on Information technology
education (pp. 195–200). New York, NY, USA: ACM.
Tordai, A., van Ossenbruggen, J., Schreiber, G., & Wielinga, B. (2011). Let’s agree to
disagree: On the evaluation of vocabulary alignment. In Proceedings of the sixth
international conference on knowledge capture (pp. 65–72). New York, NY, USA:
ACM.
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#NoessnerN10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#NoessnerN10
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0915
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0915
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0915
http://oaei.ontologymatching.org/
http://oaei.ontologymatching.org/
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#PesquitaSCC10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#PesquitaSCC10
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0945
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0945
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0945
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0955
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0955
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0955
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#QuixGKL08a
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#QuixGKL08a
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#ReulP08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#ReulP08
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0980
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0980
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0980
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0980
http://refhub.elsevier.com/S0957-4174(14)00514-4/h0980
http://dblp.uni-trier.de/db/journals/tsi/tsi28.html#SafarR09
http://dblp.uni-trier.de/db/journals/tsi/tsi28.html#SafarR09
http://dblp.uni-trier.de/db/conf/semweb/om2011.html#SchaddR11
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#SchaddR12a
http://dblp.uni-trier.de/db/conf/semweb/om2012.html#SchaddR12a
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1020
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1020
http://www.scopus.com/
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1030
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1030
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1030
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1030
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1040
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1040
http://dblp.uni-trier.de/db/books/collections/Virgilio2009.html#ShvaikoGY09
http://dblp.uni-trier.de/db/books/collections/Virgilio2009.html#ShvaikoGY09
http://www.ihtsdo.org/index.php?id=545
http://www.ihtsdo.org/index.php?id=545
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1065
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1065
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1065
http://www.cwi.nl/media/publications/wise05troncy.pdf
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#SunnaC07
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#SunnaC07
http://dblp.uni-trier.de/db/conf/pakdd/kdll2006.html#TanJLAS06
http://dx.doi.org/10.1016/j.websem.2006.06.001
http://dx.doi.org/10.1007/978-3-540-30483-8_58
http://dx.doi.org/10.1007/978-3-540-30483-8_58
http://dx.doi.org/10.1109/WI.2007.10
http://dx.doi.org/10.1109/WI.2007.10
http://dx.doi.org/10.1109/WI.2007.10
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1165
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1165
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1165
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1165
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1170
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1170
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1170
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1170
L. Otero-Cerdeira et al. / Expert Systems with Applications 42 (2015) 949–971 971
Trojahn, C., Quaresma, P., & Vieira, R. (2012). Exploiting majority acceptable
arguments for ontology matching. International Journal of Artificial Intelligence, 8,
1–19.
Tyl, P., & Loufek, J. (2009). Comp – comparing ontology matching plug-in. In Fifth
international conference on next generation web services practices, 2009.
NWESP’09 (pp. 44–49).
Vouros, G. A., & Kotis, K. (2005). Extending hcone-merge by approximating the
intended meaning of ontology concepts iteratively. In A. Gómez-Pérez & J.
Euzenat (Eds.), ESWC (pp. 198–210). Springer. URL: .
Wang, P. (2011). Lily results on SEALS platform for OAEI 2011. In P. Shvaiko, J.
Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM, CEUR-WS.org. (pp.
156–162).
Wang, P., & Xu, B. (2007). Lily: The results for the ontology alignment contest oaei
2007. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org.
URL:.
Wang, P., & Xu, B. (2008). Lily: Ontology alignment results for oaei 2008. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. URL:.
Wang, P., & Xu, B. (2009). Lily: Ontology alignment results for oaei 2009. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal
(Eds.), OM, CEUR-WS.org. URL: .
Wang, Z., Zhang, X., Hou, L., Zhao, Y., Li, J., Qi, Y., & Tang, J. (2010). RiMOM results for
OAEI 2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao,
& I. F. Cruz (Eds.) OM, CEUR-WS.org. (pp. 195–202).
Warin, M., & Volk, H.M. (2004). Using WordNet and semantic similarity to
disambiguate an ontology. Technical Report. STOCKHOLMS UNIVERSITET
Institutionen för lingvistik.
Wennerberg, P. (2009). Aligning medical domain ontologies for clinical query
extraction. In Proceedings of the 12th conference of the european chapter of the
association for computational linguistics: student research workshop, Association
for Computational Linguistics, Stroudsburg, PA, USA (pp. 79–87).
WordNet (2013). WordNet. URL: .
Wrigley, S. N., García-Castro, R., & Nixon, L. (2012). Semantic evaluation at large
scale (seals). In Proceedings of the 21st international conference companion on
World Wide Web (pp. 299–302). New York, NY, USA: ACM.
Xu, P., Tao, H., Zang, T., & Wang, Y. (2008). Alignment results of sobom for oaei 2009.
In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A.
Rosenthal (Eds.), OM, CEUR-WS.org. URL: .
Xu, P., Wang, Y., Cheng, L., & Zang, T. (2010a). Alignment results of SOBOM for OAEI
2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao, & I. F.
Cruz (Eds.), OM, CEUR-WS.org. (pp. 203–2011).
Xu, P., Wang, Y., Cheng, L., & Zang, T. (2010b). Alignment results of sobom for oaei
2010. In P. Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, M. Mao, & I. F.
Cruz (Eds.), OM, CEUR-WS.org. URL: .
Xu, P., Wang, Y., & Liu, B. (2012). A differentor-based adaptive ontology-matching
approach. Journal of Information Science, 38, 459–475.
Yaghlane, B. B., & Laamari, N. (2007). OWL-CM: OWL combining matcher based on
belief functions theory. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.),
OM, CEUR-WS.org. (pp. 206–218).
Zargayouna, H., Safar, B., & Reynaud, C. (2007). Taxomap in the oaei 2007 alignment
contest. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He (Eds.), OM,
CEUR-WS.org. URL: .
Zghal, S., Kachroudi, M., Yahia, S. B., & Nguifo, E. M. (2009). Oacas – ontologies
alignment using composition and aggregation of similarities. In J. L. G. Dietz
(Ed.), KEOD (pp. 233–238). INSTICC Press. URL:..
Zghal, S., Kachroudi, M., Yahia, S. B., & Nguifo, E. M. (2011). OACAS: Results for OAEI
2011. In P. Shvaiko, J. Euzenat, T. Heath, C. Quix, M. Mao, & I. F. Cruz (Eds.), OM,
CEUR-WS.org. (pp. 190–196).
Zghal, S., Yahia, S. B., Nguifo, E. M., & Slimani, Y. (2007). SODA: An OWL-DL based
ontology matching system. In P. Shvaiko, J. Euzenat, F. Giunchiglia, & B. He
(Eds.), OM, CEUR-WS.org. (pp. 261–267).
Zhang, H., Hu, W., & Qu, Y. (2011). Constructing virtual documents for ontology
matching using mapreduce. In J. Z. Pan, H. Chen, H. G. Kim, J. Li, Z. Wu, & I.
Horrocks, et al. (Eds.), JIST (pp. 48–63). Springer.
Zhang, J., Lin, P., Huang, P., & Wu, G. (2011). Research on semantic web service
composition based on ontology reasoning and matching. Communications in
Computer and Information Science CCIS, 243, 450–457.
Zhang, S., & Bodenreider, O. (2007). Hybrid alignment strategy for anatomical
ontologies: Results of the 2007 ontology alignment contest. In P. Shvaiko, J.
Euzenat, F. Giunchiglia, & B. He (Eds.), OM, CEUR-WS.org. (pp. 1–11).
Zhang, X., Zhong, Q., Li, J., & Tang, J. (2008). Rimom results for oaei 2008. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, & H. Stuckenschmidt (Eds.), OM, CEUR-
WS.org. URL: .
Zhang, X., Zhong, Q., Shi, F., Li, J., & Tang, J. (2009). Rimom results for oaei 2009. In P.
Shvaiko, J. Euzenat, F. Giunchiglia, H. Stuckenschmidt, N. F. Noy, & A. Rosenthal
(Eds.), OM, CEUR-WS.org. URL: .
Zhu, J. (2012). Survey on ontology mapping. Physics Procedia, 24(Part C), 1857–1862.
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1175
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1175
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1175
http://dblp.uni-trier.de/db/conf/esws/eswc2005.html#VourosK05
http://dblp.uni-trier.de/db/conf/esws/eswc2005.html#VourosK05
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#WangX07
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#WangX08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#WangX08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#WangX08a
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#WangX08a
http://wordnet.princeton.edu/
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1230
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1230
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1230
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#XuTZW08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#XuTZW08
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#XuWCZ10
http://dblp.uni-trier.de/db/conf/semweb/om2010.html#XuWCZ10
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1250
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1250
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#ZargayounaSR07
http://dblp.uni-trier.de/db/conf/semweb/om2007.html#ZargayounaSR07
http://dblp.uni-trier.de/db/conf/ic3k/keod2009.html#ZghalKYN09
http://dblp.uni-trier.de/db/conf/ic3k/keod2009.html#ZghalKYN09
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1280
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1280
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1280
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1285
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1285
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1285
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#ZhangZLT08
http://dblp.uni-trier.de/db/conf/semweb/om2008.html#ZhangZLT08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#ZhangZSLT08
http://dblp.uni-trier.de/db/conf/semweb/om2009.html#ZhangZSLT08
http://refhub.elsevier.com/S0957-4174(14)00514-4/h1305
Ontology matching: A literature review
1 Introduction
2 Procedures
3 Statistical results
3.1 Articles sorted by publication year
3.2 Articles sorted by data source
4 Classification
4.1 ‘Reviews’ category
4.2 ‘Matching Techniques’ category
4.3 ‘Matching Systems’ category
4.4 ’Processing Frameworks’ category
4.5 ‘Practical Applications’ category
4.6 ‘Evaluation’ category
5 Limitations of the Literature Review
6 Trends in Ontology Matching: Practitioner-oriented Survey
6.1 Participants and Survey design
6.2 Survey results
6.2.1 Background questions
6.2.2 Research field questions
6.2.3 Future challenges questions
7 Limitations of the Survey
8 Conclusions
References