key: cord-0078018-fgswzlwp
authors: Liu, Tao; Yu, Zhongyang
title: The relationship between open technological innovation, intellectual property rights capabilities, network strategy, and AI technology under the Internet of Things
date: 2022-05-11
journal: Oper Manag Res
DOI: 10.1007/s12063-021-00242-8
sha: 62f5bf9395368a70f8402aa9419cae7ab7e68e26
doc_id: 78018
cord_uid: fgswzlwp

The problems faced by the open technological innovation of China’s new ICT (information and communications technology) industry under IoT (Internet of Things) technology are expected to be analyzed to improve the overall innovation ability and ensure the sustainable development of related industries. An evaluation model is constructed for open technological innovation in the IoT industry by analyzing the development of IPR (Intellectual Property Rights) management, network strategy, and AI (artificial intelligence) technology under the development of IoT technology. Meanwhile, IBM SPSS Statistics 20.0 and IBM SPSS Amos 19.0 are used to analyze the data information of 306 enterprises in the information technology industry. Besides, the proposed hypotheses are verified by factor analysis, multiple regression, and Back Propagation Neural Network. Finally, a new evaluation index system is constructed for open technological innovation in the new ICT industry. The development of IoT technology provides a primary guarantee for the open technological innovation of the new ICT industry, and the network strategy has the greatest influence on the internal knowledge output mode. Besides, the experimental results indicate that the IoT and artificial intelligence have a critical display value for the open technological innovation of the emerging ICT industry, with the highest weight ratio of 48.25%. This result demonstrates that artificial intelligence is positively correlated with the external input information. Intellectual property management is a crucial guarantee of open technology innovation in the ICT industry. The evaluation model of open technological innovation in the ICT industry has good performance through case analysis, with the highest accuracy of 91.25%. Therefore, the evaluation index system reported here can reflect the important factors affecting the development of innovative technology, which can provide a theoretical basis and practical value for improving the existing open technology innovation system.

With the advent of the era of the knowledge economy, knowledge has gradually become the most important and influential factor in productivity. Knowledge economy brings information and sustainable innovation. In addition, knowledge has gradually become the dominant capital and plays an increasingly prominent role in enterprise decision-making. Therefore, knowledge-based open technological innovation enterprises occupy a more important position in China's economy. At present, innovation is becoming more and more important, and innovation management has become the key of enterprises. Innovation plays a vital role in national economic development and enterprise growth. Especially in the competitions, the enterprises that can innovate, transform innovative thinking into results, use this knowledge to transform their business processes, products, or services, improve and develop new business fields and new markets will have better market value and growth potential than other enterprises. Therefore, under the background of the Internet of Things (IoT) and artificial intelligence (AI), it is of great significance to study the impact of intellectual property management capability on open technological innovation. Lin et al. (2021) firstly used the data packet analysis window with ideal window width to dynamically investigate the technological innovation efficiency of China's high-tech industry regarding the province, region, and industry from 2009 to 2016. The results showed that the efficiency of the high-tech industry was low, showing a waveform trend, and there were great differences between provinces and regions. The efficiency of the eastern region was always the highest, while that of the northeast region was the lowest. In addition, the efficiency of the eastern and western regions showed a downward waveform trend, and the efficiency of the central and northeast regions presented an upward waveform trend. The efficiency of each sub-industry had significant differences between regions and provinces, and the efficient distribution of each province in the five sub-industries was different. In the whole evaluation period, no province has always been at the forefront of innovation. Arogundade et al. (2020) pointed out that security risk management was a knowledge-intensive program, which needed to monitor and capture relevant information to help managers make correct decisions. They proposed a syntax reinforcement model of security management during the life of an information system. The model supports the continuous collection of identified threat behaviors from intrusion detection systems, filtering and analyzing threats in time snapshots, and re-evaluating the security countermeasures, including security administrator (S-Admin), managers, and security management system as stakeholders. Hu et al. (2021) pointed out that during the COVID-19 crisis, export-oriented manufacturing companies worldwide, especially manufacturing enterprises in developing countries, were forced to lay off workers and deal with problems related to value distribution. Serious survival problems led to major challenges related to corporate social responsibility. Although limited research has discussed relevant issues in the non-Western context, the research adopts the global perspective of business model and transaction cost theory to fill this gap by investigating the mechanisms of different dimensions such as enterprise safety report implementation, corporate performance, and value distribution. According to the sample survey of Chinese listed manufacturing enterprises, the technical dimension of enterprise safety report is negatively correlated with enterprise performance, the content dimension of an enterprise safety report is positively correlated with enterprise performance, and the value distribution actively regulates the relationship between all three enterprise safety dimensions and enterprise performance. Alcaide Muñoz and Rodríguez Bolívar (2021) pointed out that cities are developing strategies to meet the complex challenges of global change and sustainability. These strategies include the implementation of information and communications technology (ICT) as a good driver for sustainability, as digital transformation can promote sustainable development strategies and provide opportunities to accelerate transformation. Pustovrh et al. (2020) proposed to use the open innovation paradigm to analyze the development of the entrepreneurship support ecosystem. They held that the enterprise accelerator played a key role in the emerging entrepreneurial ecosystem. Their analysis used qualitative data from business accelerators in this context. They suggested that public policies should support the open innovation activities of key actors in the entrepreneurial ecosystem in an environment where the entrepreneurial support ecosystem is underdeveloped. Through the use of open innovation, the accelerator has established a wider network of relationships with actors outside the system, thereby increasing the capacity within the system and embedding it into the global innovation system. Hence, exploring open technological innovation of the new ICT industry has important practical value for improving innovation ability and deepening reform and opening up.

To sum up, there is an urgent need to solve the problems faced by the open technological innovation of the new ICT industry under the IoT technology to provide the overall innovation ability of enterprises. This paper puts forward the evaluation index system of open technological innovation in the ICT industry by searching the relevant literature. The system integrates its characteristics and the actual situation. Factor analysis, multiple regression, and backpropagation neural network (BPNN) is selected to assign weights to the indicators. The innovation lies in the practical evaluation from intellectual property management, network strategy, and AI technology. The advantage of the model design is to explore the impact of these three factors on employee behavior from the knowledge management capability, network strategy, and AI technology, which is very important for the future research of open technological innovation. In addition, it also analyzes the impact of different management strategies and technologies from multiple dimensions, which provides theoretical support for improving relevant theories.

The paper is structured as follows. Section 1 mainly introduces the research background and significance and explains the research value. Section 2 summarizes the advantages and disadvantages of relevant research and highlights the research significance and value. Section 3 expounds on the relevant theoretical knowledge of open technological innovation and IoT technology and designs a questionnaire on them. Section 4 provides the results of the questionnaire and carries out regression analysis. Section 5 discusses the experimental results to understand the correlation between the current open technology innovation and the IoT technology and puts forward corresponding suggestions. Finally, Sect. 6 summarizes the research methods and results, highlights the shortcomings, and analyzes the follow-up development.

The most important thing for studying open technological innovation in the new information technology industry is establishing an open technological innovation evaluation system, which can evaluate enterprises in all aspects, further create a positive innovation environment, and effectively improve the efficiency and quality of enterprises' innovation results on innovation platform (Yan et al. 2020 In open technological innovation, emerging technologies represented by IoT have become the key to national technological innovation. Many scholars have reported the impact of IPR management on open technological innovation. This will easily lead to the negative impact of excessive openness on organizational performance. As for the impact of AI technology on open technological innovation, studies have shown that AI can greatly improve the efficiency of the existing economy, which may lead to a shift from more conventional labor-intensive research to passively generated large datasets and enhanced prediction algorithms (Wu et al. 2020a ). Khan et al. (2020) provided insights on the support of the IoT and the latest development of emerging technologies and considered the challenges that enterprises might encounter in innovative development from supportive technologies, applications, business models, and social and environmental impacts. 

Open technological innovation comes from traditional closed innovation. Its essence is based on the open internet, the integration of resources, and the use of the group's ability to overcome difficulties; its goal is to reduce the risk of innovation and timely transform the results of innovation (Xu et al. 2020) . Furthermore, open technological innovation can fully utilize and integrate external resources. Therefore, open technological innovation can be considered as the process of the inflow and outflow of resources and the final realization of commercialization (Lv et al. 2021a ). Specifically, the model of open technological innovation is shown in Fig. 1 . As shown in Fig. 1 , the open technological innovation model's operating mechanism is that there are two necessary conditions for enterprises to implement open technological innovation: (1) enterprises need resources in terms of technology, manufacturing, and market. These resources are the foundations of enterprise innovation and are indispensable.

(2) Using the above three resources by the enterprise is based on enterprises' research and development system, production system, and sales system. These three systems are interdependent. The significant connection and management of these systems is a prerequisite for practical innovation. Figure 2 presents a further analysis of the open technological innovation mechanism. Figure 2 suggests that enterprises require to acquire and fully utilize endogenous and external innovation resources in the operating mechanism of the open technological innovation model. Key customers, suppliers, and competitors constitute the principal channels of external innovation resources for enterprises. Enterprises continuously acquire external resources through channel management and maintenance and strengthen internal innovation to generate more internal creativity (Ruhrmann et al. 2021 (Yuan and Gasco-Hernandez 2021) .

With the continuous emergence of disruptive technologies and the accelerated market structure transformation, it is impossible to achieve revolutionary innovation relying solely on internal resources and forces of a single subject. Traditional closed innovation and "1 + N" open innovation are facing many challenges. With higher openness and stronger interaction, "N + N" sharing innovation comes into being. Figure 3 indicates the specific structure of open technological innovation based on IoT. The core and foundation of IoT is still the internet, which is an extension and expansion of the internet. Clients have expanded information exchange and communication to the connection between any two objects. Therefore, according to the agreement, IoT connects everything to the internet for information exchange and communication through information sensing equipment such as radio frequency identification, infrared sensors, 

IPR management refers to the administrative and judicial activities carried out by national departments to ensure the implementation of the intellectual property legal system and safeguard the legitimate rights and interests of intellectual property owners. Intellectual property is the core competitiveness of an enterprise. Therefore, resources of intellectual property are the first strategic resource. Strategy implementation: it is a systematic and forward-looking review of the development and operation of intangible assets, such as IPR. Feasible mid-and long-term intellectual property development plans and goals are formulated according to the enterprise's development goals, operating conditions, competitive landscape, and market development trends (Vecchiato 2020) . Previous reports have taken the strategy of intellectual property as an essential measurement indicator for evaluating an enterprise's innovation capability (Stanisławski and Szymonik 2021; Jiang et al. 2021) . Information system: it tracks and predicts technological trends, industry trends, competitive areas, and market trends, providing a decision-making basis for formulating the enterprise's research and development plan, thereby determining the enterprise's product development goals. Ferraris et al. (2020) established the correlation between the knowledge search system and the evaluation of enterprise innovation; compared with the Chinese market and enterprises, the correlation between the knowledge search system and the open technological innovation of international market and enterprises is stronger. Research and development of IPR: as an intangible asset resource, it should be scientifically and reasonably developed, utilized, and operated to commercialize and capitalize on the enterprise's knowledge resources and IPR. Therefore, the contribution rate of IPR in the enterprise's new assets is increased, becoming new profit growth points. Grimaldi et al. (2021) found that IPR protection capabilities have different impacts on the efficiency of open technological innovation of enterprises on different scales. According to the above results, the following hypotheses are proposed to test the relationship between the enterprise's capabilities of IPR management and technological innovation. First, protection of IPR: only by obtaining patent protection can it finally form its unique market competitive advantage. The contribution of intellectual property rights to the new assets of enterprises has increased and become a new profit growth point. The world powers attach great importance to patent protection, so the world economic and technological powers are also very strong in patents. 

Network strategy is the ability to evaluate the network situation and environment from a strategic level and dynamically plan and change the degree of openness based on particular situations. Network strategy arises from the interdependence of the environment. These interdependencies come from many aspects. The awareness of interdependence may come from external threats faced by countries, regions, and enterprise clusters Mi et al. 2021 ). Network process: it is the actual processing capability when the process is in a stable state. Analyzing the process capability of the processing process can help understand the quality assurance capability of each product in the manufacturing process at any time, thereby providing the necessary information and basis for ensuring and improving product quality (Saunila 2020) . Research has found that enterprises may limit their ability to improve their organizational performance from open technological innovation due to various factors, such as their limited absorptive capacity, the selection of issues during an open period, and different levels of focuses on some aspects of the enterprises at a particular stage. It is easy to cause adverse effects on organizational performance due to excessive openness (Sun et al. 2020; Lv et al. 2021b) . Network knowledge connects technology and humans to effectively combine intellectual capital, structural capital, and customer capital. Enterprise network relationship refers to a long-term association between enterprises established by a group of independent and interrelated enterprises and various institutions for common goals based on the specialized division of labor and collaboration (Buchnea et al. 2020; Yang and Liu 2020; Lv et al. 2020a) proposed that the relationship between an enterprise's network capabilities plays a vital role in solving the enterprise's innovation capabilities, which is the basis of collaborative innovation of enterprises. Based on the above analysis, the following hypotheses are proposed: 

AI is the production of a new intelligent machine that can react similarly to human intelligence. Research in this field includes robotics, language recognition, image recognition, natural language processing, and expert systems (Harré 2021; Lv et al. 2020b) . AI can effectively influence the technological innovation activities of enterprises. Therefore, AI is seen as a critical variable affecting innovation. Enterprises with AI technology can better identify and seize rapidly changing market opportunities and adjust their behaviors according to market needs, thereby responding to market changes. Many enterprises seek, learn, and use the knowledge required by AI, encouraging enterprises to develop new products faster and increase the speed of innovation (Jumani 2021; Lv et al. 2021c) . In this process, the innovation performance of the enterprise is improved, and at the same time, AI will have a self-reinforcing effect, enhancing the ability of the enterprise to acquire and learn knowledge in the future. Here, it is believed that enterprises affect open technological innovation through four aspects: acquisition, digestion, transformation, and utilization of AI (Wu et al. 2020b Figure 4 presents the framework of the hypotheses.

BPNN is commonly utilized for training ANNs (Artificial Neural Networks) in combination with optimization methods. This method calculates the gradient of the loss function for all weights in the network. This gradient will be fed back to the optimization methods to update the weight, thereby minimizing the loss function. The particular structure is shown in Fig. 5 , including the input layer (I), the hidden layer (H), and the output layer (O). In BPNN, the number of hidden layer nodes is uncertain. The specific calculation is as follows:

In (1), h represents the number of nodes in the hidden layer, m represents the number of nodes in the input layer, n represents the number of nodes in the output layer, and α represents the adjustment constant. The process of forwarding transmission: if the weight between node i and node j is w ij , the

threshold of node i is x i , and the threshold of node j is b j . The output value of each node is determined based on the output value of all nodes in the upper layer, the weight of the current node and all nodes in the upper layer, the threshold value of the current node, and the activation function f. The specific calculation process is as follows:

In BPNN, the reverse transmission sub-process of error signals is more complicated based on Widrow-Hoff learning rules. For example, if all the results of the output layer are d j , the error function will be: The learning rule is to continuously adjust the weights and thresholds of the network along the direction of the fastest descent of the relative error square sum. According to the gradient descent method, the correction of the weight vector is proportional to the gradient of E(w,b) at the current position and η is a constant. For the j-th output node:

According to the BPNN model, from the perspectives of IPR management, network strategy, and AI technology, the specific indicators are shown in Table 1 . Finally, an evaluation system for open technological innovation of the new ICT industry is established, as shown in Fig. 6 . Figure 6 reveals that the BPNN model is applied to the performance evaluation of enterprise open technology innovation, including data processing, network design, detailed design, and other steps and links. Furthermore, each link is closely integrated and interlocked to form an effective entirety, ensuring the smooth development of the model in the performance evaluation of enterprises' open technological innovation.

(1) Data processing: the first is factor analysis processing. The principal component analysis (PCA) of factor analysis is necessary to convert the high-dimensional data into low-dimensional data based on the original information and effectively extract the common factors of the data. The constructed evaluation system of open technological innovation performance has many indicators affecting the operational efficiency of BPNN and the evaluation results. Therefore, the advantages of factor analysis are fully utilized to analyze the links between indicators effectively. First, the training samples of the BPNN are confirmed as dimensionality reduction

data, and a total of 14 comprehensive factors are obtained. Simultaneously, it is necessary to obtain the required data after the indicators are determined through reliable network platforms, such as Shenzhen GTA Education Tech Ltd. (GTAFE), Zhejiang Hithink Royal Flush Information Network Co., Ltd., and cninf run by Shenzhen Securities Information Co., Ltd. Finally, the performance of listed companies is evaluated by factor analysis, and the output data of training BPNN are the evaluation results. Second, the data are standardized. Data standardization is basically divided into several steps, including development, candidate, approval or rejection, and archiving. It is a process of standardizing the definition, organization, supervision, and protection of data. There are many evaluation indicators for the open technological innovation performance of enterprises. If financial or non-financial data are utilized directly for analysis, the evaluation results will not match the actual results due to the large differences between the indicators and the inconsistency of the dimension units. Therefore, it is necessary to eliminate the influence caused by the indicator dimension through standardization, reduce the error caused by the index difference, and improve the convergence efficiency of BPNN and the accuracy of performance evaluation results.

(2) Network design: The system randomly assigns the BPNN an initial weight, and the expected value is compared with the output result calculated through multiple cycles of the assignment. The error is effectively controlled within the preset value range. Notably, the design includes selecting target data and input data; the network parameters are initialized through the established BPNN model; the computer automatically completes the construction of the network training sample data and loops to the preset value; the network training is completed, the threshold and weight are confirmed, and then the final evaluation result of the time limit is output. (3) Detailed design: design of the network layers' number: a three-layer neural network can express any nonlinear mapping problem. Therefore, it is used as the basis for performance evaluation to solve the problem of being too complicated and falling into a local solution. Node: indicator in the evaluation indicator system is optimized by factor analysis. The input node of the BPNN model is determined as the obtained factor indicator, which can comprehensively show the factor indicator of the overall information. According to literature and experiments, the number of layer nodes is set as 1. Parameter selection: BPNN budget accuracy and convergence efficiency are affected by the selection of threshold and weight. The maximum number of iterations should be set based on GA, which is determined as 1000 times here to avoid the defect of random BPNN assignment, fully utilize the global search advantage of GA, make the initial threshold and weight close to the final value, and avoid the problem that the convergence has not been achieved after the finite iterations of the BPNN. The function from the input layer to the output layer uses the Tansig function corresponding to the S type, the output layer uses the Purelin linear function, the learning rule uses the Traingdx function, the performance evaluation uses the Mes function, the model number is set to 1000, the accuracy is set to 0.0001, and the rest are all the default parameters of the system.

Principles of questionnaire design: the questionnaire method has been widely used in modern management research. Compared with other research methods, the questionnaire method has the characteristics of low cost, less time-consuming, and a large sample size. Especially with the development of information communication, the questionnaire method has been more widely promoted. However, the prominent problem of the questionnaire method is that the questionnaire may be constructed randomly and misused. Based on this, the questionnaire design should follow some principles. In research methodology for management, Li put forward the following principles: first, the questionnaire should be concise, easy to answer, and attractive, and it should avoid evasive and lengthy questions. Second, the question should be specified and correspond to a single concept. It should not contain double meanings to avoid being difficult for the respondents to answer. Third, the words used in the questionnaire should be neutral, and no leading sentences should be used to induce the respondents. Fourth, questions that are difficult to reflect authenticity should be avoided. Fifth, preconceptions should be eliminated. At the beginning of the questionnaire design, unconfirmed things should not be taken as presuppositions, making it difficult for respondents to answer. The design of the questionnaire strictly abides by the above principles and adheres to the conciseness, scientificity, and principle of the questionnaire.

Procedures of questionnaire design: first, current literature in China and other countries has been consulted to look for the relevant variable measurement scale. On the basis of existing researches, the preliminary dimension division of the measurement variables is formed according to the actual development of Chinese enterprises. Knowledge absorptive capability is divided into four dimensions: knowledge acquisition ability, knowledge digestion ability, knowledge transformation ability, and knowledge utilization ability. Because the concept of open technological innovation has been put forward for a short time, the research on the open technological innovation model is not enough, and the research on its dimension division is few. Based on the existing qualitative research literature on the open technological innovation model, two dimensions are created: the external knowledge input and the internal knowledge output models. The innovation performance is not divided into further dimensions, but the intensity of enterprise research and development investment and market development are adopted as measurement dimensions, and the evaluation indexes of enterprise patent, product development, and market and business development are the main focuses.

Subsequently, the final scale is formed through field investigation, interviews, and academic discussion. Firstly, seriously consider the rationality and clarity of the questionnaire. Secondly, through the on-site discussion with the middle-level and above managers of the target enterprise, modify and supplement the expression and content of the questionnaire. Thirdly, combined with the opinions of professionals in relevant research, revise the questionnaire and finally form a reasonable and scientific questionnaire.

Contents of questionnaire design: the questionnaire is composed of two parts. The first part involves the basic information of the respondents and the enterprise. The second part is the survey of enterprise IPR management, network strategic ability, AI technology evaluation and open technological innovation of the new ICT industry. It is divided into four dimensions with 13 questions. The third part is the open technological innovation mode survey, which is divided into two dimensions with 7 questions. The fourth part is the measurement of enterprise innovation performance, with 6 questions. Thus, the questionnaire consists of 26 questions. It adopts the Likert scale of five points to measure the result. The points from 1 to 5 are divided into five options, which indicate "very disagree", "relatively disagree", "generally agree", "relatively agree" and "strongly agree" [33] . In this survey, online questionnaires are distributed to different industries and companies. From October 2018 to October 2019, 352 survey responses were received from 1,580 new ICT companies registered in the China Software Industry Association. After excluding questionnaires with obvious problems, there remain 306 questionnaires for analysis. In this questionnaire survey, the whole process of questionnaire design, distribution, and data collection does not involve the personal privacy of the respondents; the whole questionnaire survey is conducted with the consent of participants (not less than 18 years old); the questionnaire is not open to the public, only for research purposes.

A questionnaire survey was performed to analyze the impact of IPR management capabilities, network strategic capabilities, and AI on open technological innovation. The reliability and validity of the data should be analyzed, in which Cronbach's alpha tests the reliability of the questionnaire. The calculation process is as follows:

In (1), K is the number of items on a particular scale (some data are misrepresented as the number of samples), σ 2 X is the variance of the total sample (variance of the total score of each item on a particular scale), and σ 2 Y is the variance of the observed sample (variance of the scores of each subject in a particular item). Generally, if Cronbach's alpha coefficient is above 0.7, the reliability of the survey data is excellent .

As for validity analysis, Kaiser-Meyer-Olkin (KMO) and Bartlett sphere test methods are used to test whether variable samples can be factored. All the data are input into SPSS 20.0 and AMOS 19.0 software for the validity test of the variable sample factors. The specific process refers to López et al. (2020) . Also, PCA and maximum variance orthogonal rotation are utilized to observe each factor's high loads and classification. Generally, if KMO is greater than or equal to 0.7 and the load coefficient of each item is greater than 0.5,

� the combination of factors between items of the same variable will be realized (Rivero-Antunez et al. 2020).

First, the regression coefficient point's estimated value can be obtained (different samples obtain different values) by substituting all the data sample values, thereby obtaining the multiple linear regression equation of the samples. In this equation, under the premise that the model and data satisfy the basic assumptions, parameter estimation can be obtained by the least square estimation. The equation is as follows:

In (2), S 2 e is the variance of a random variable, k is the number of independent variables, Y is the value of the dependent variable obtained from the experiment, and Y i is the estimated value of the variable point obtained by the regression equation. By testing the correlation between the variables and the size of the correlation coefficient, the rationality of hypothesis verification and theoretical models can be preliminarily judged. Due to the massive amount of data involved, the SPSS software is employed to test the simple correlations among the explained variables, explanatory variables, and intermediate variables according to the correlation coefficient (Rivero-Antúnez et al. 2021). Figure 7 shows the basic information of all the enterprises surveyed. The objects of this survey are mainly new ICT enterprises, which involve six types of new ICT industries. The results show that the distribution of the establishment time of the surveyed enterprises is average, with the largest number of enterprises established 10-20 years, accounting for 36.1%. The scale of the surveyed enterprises is mainly concentrated in "200-500 staff" and "more than 500 staff", accounting for 33.2% and 27.8%, respectively. The industry distribution of the surveyed enterprises is even. The number of enterprises in the primary industries accounts for 96.7%, of which the proportion of the cloud computing industry is the largest, accounting for 20.3%, followed by the electronic information technology industry, accounting for 19.1%. The number of enterprises with an annual operating income of 5-10 million CNY is the largest, accounting for 34%. It shows that the results of this survey cover a wide range of industries and involve many enterprises, which can better explain actual problems. Figure 8 shows the results of reliability and validity tests for all the data of the questionnaire. The specific reliability statistics are obtained through SPSS software, as shown in Fig. 7 . As for the reliability of the variable data, Cronbach's alpha coefficient is greater than 0.7; hence, the reliability is high. Table 2 shows that after the KOM test, all the values are also greater than 0.7, and Sig. is 0.000. Thus, the above analysis results show that there is no strong correlation among the measurement items in the survey, and the survey has good discrimination validity. Figure 9 shows the relationship between all the factors by using multiple regression analysis. First, the relationship between IPR management capabilities and enterprises' innovation performance is tested. The results show that under P = 0.05, the four dimensions of IPR management capability, i.e., property strategy implementation capability, property rights information system, property rights research and development capability, and property rights protection capability, have a significant positive correlation with the innovation performance of enterprises. Hence, hypotheses H1, H1a, H1b, H1c, and H1d are true. Therefore, the results show that an enterprise's ability to absorb knowledge from four aspects will improve its innovation performance. Second, the relationship between network strategic capabilities and enterprises' innovation performance is examined. The results show that under P = 0.05, the two dimensions of the open technological innovation model, i.e., network process capability and network knowledge capability, significantly correlate to enterprises' network relationship capability and enterprises' innovation performance. Hence, hypotheses H2, H2a, H2b, and H2c are true. Therefore, the ability of enterprises to implement network strategies will improve their innovation performance. Finally, the relationship between AI and open technological innovation models is examined.

(7) S 2 e = ∑ � Y − Y i � 2 n − k − 1 fijmh

The results show that AI is significantly positively correlated with input patterns of external knowledge and has no significant correlation to the output patterns of internal knowledge. AI digestibility is significantly positively correlated to the input patterns of external knowledge and the output patterns of internal knowledge. There is no significant correlation between AI transformation ability and the input patterns of external knowledge and the output patterns of internal knowledge. On the other hand, AI utilization ability is significantly positively correlated to the input patterns of external knowledge and the output patterns of internal knowledge. Therefore, hypotheses H3b and H3d are valid; hypotheses H3a and H3c are invalid. Tables 3 and 4 (Wu et al. 2020a) . The external knowledge input model of IPR management capability has a more significant impact on innovation performance than the internal knowledge output model. A possible reason is that open technological innovation is an innovation model based on the evolution of excellent Western enterprises and technological environments. The internal knowledge output model is a higher-level model and is a model adopted by large enterprises. China still stays in the stage of technology accumulation. The stage goal of enterprises is to accumulate innovative resources and introduce external innovative resources as much as possible to achieve technological breakthroughs. They have not yet reached the stage of technology output or internal technology commercialization. However, large enterprises account for a small share of the sample data, which may not effectively demonstrate this model. The overall performance of the constructed performance model is higher than that of the traditional model, up to 91.25%. The above results also further illustrate the effectiveness of the model proposed.

The empirical study indicates that in the correlation analysis, the four aspects of IPR management capability and innovation performance have a significant positive correlation; in the regression analysis, the knowledge absorptive capability also significantly affects innovation performance. This is consistent with the basic hypotheses and in line with the research results of scholars in China and other countries. Zahra and George theoretically analyzed the impact of the four dimensions of knowledge absorptive capacity on innovation performance. They believed that the four dimensions of absorptive capacity could elevate the speed of enterprise resource acquisition, improve the direction of resource acquisition and utilization, enhance the cognitive ability of employees, perfect the mechanism of knowledge storage and flow within the enterprise, and ultimately promote the enterprise performance. The research on the innovation of the high-tech industry confirms that there is a significant positive correlation between knowledge absorptive capacity and innovation performance of enterprises, which is consistent with the research of scholars in China and other countries. The other three capabilities play a major role in the understanding, digestion, and integration of knowledge in the organization, and they do not have the obvious support of knowledge innovation entities. Thus, they have little effect on innovation performance, which is consistent with the research results of some scholars. The relationship and mechanism between intellectual property management, network strategy, artificial intelligence technology, and enterprise innovation performance are discussed in the present work. According to the obtained results, the following suggestions are made. First, AI is a critical technology for future information development. Enterprises need to cultivate professionals in AI-associated fields, implement developmental innovation models as soon as possible, make more active use of all internal and external resources, increase investment in AI technology, increase Technological innovation is the key driving force for the development of an industry. The intelligent service system in the IoT environment will become an important infrastructure for future society. As the key technological innovation of IoT, the intelligent service system coordinates maps the real physical space and virtual information space to realize the integration of communication, calculation, and control. Besides, an intelligent service system enables active collaborative interaction between objects and between people and objects in a new way, thus weaving an intelligent collaborative network with endogenous interconnection in the physical world. Under the current macroeconomic situation, China's economy has entered a period of structural slowdown. The traditional way of relying on intensive investment and demographic dividends to promote economic growth is unable to adapt to China's future economic development. China needs to seek a new growth point of the economy from technological innovation and high-quality investment in the future. At this time, the development of the AI industry provides a good opportunity for China to stabilize economic growth and realize overtaking on corners. AI is at the peak of the development wave. The rise of this wave is mainly due to the leap of data, computing power, and algorithm. The first is the big data explosion brought by the popularization of the mobile internet; the second is the leap of computing power and the continuous decline of computing costs brought by the application of cloud computing technology; the third is the application and promotion of machine learning on the internet. However, many experts in the industry believe that there is no possibility of transformation from AI to general AI in mechanism, and the large-scale commercial application of AI will still be a long-term and tortuous process.

With the continuous development and improvement of computing infrastructure, deep learning algorithms, and big data technology, artificial intelligence has gradually broken through the technical bottleneck and ushered in the third rapid development period. With the unceasing progress of artificial intelligence technology, its information acquisition ability and information processing speed are improved. Consequently, the burden of the human brain and physical strength is reduced, work efficiency is enhanced, and the quality of human life and education level is improved. In the breakthrough of education mode, new information technology with artificial intelligence as the core balances educational resources, improves teaching efficiency, and optimizes the learning experience. In addition, the internet plus artificial intelligence model also makes education gradually personalized and promotes the breakthrough of the traditional education model.

In conclusion, knowledge absorptive capacity affects innovation performance significantly. Knowledge absorptive capacity is a kind of core dynamic capability for enterprises. It is the core competitive advantage that enterprises gradually cultivate through organizational learning in order to adapt to knowledge diversity and innovation uncertainty. In today's diversified and dynamic market environment, enterprises with higher knowledge absorptive capacity will better use innovation resources and carry out innovation activities flexibly and effectively.

To facilitate the innovative development of open technologies of enterprises, the open technology innovation evaluation model for the ICT emerging industry based on IoT is constructed using BPNN and multiple hierarchical regression model, from three dimensions of intellectual property, network strategy, and artificial intelligence. Through the Fig. 9 Results of multiple regression analysis. A Annva of multiple regression analysis; B model summary of multiple regression analysis performance evaluation of the relevant data and information of the actual enterprises, the model can evaluate the innovation ability of the relevant enterprises in the new ICT industry with better performance than other models. The paper logically analyzes the relationship between AI, intellectual property management, network strategy, and open enterprise technology innovation performance.

The following suggestions are proposed accordingly. Firstly, IoT is the key technology of future information development. Thus, enterprises need to cultivate talents in related fields to implement open technological innovation as soon as possible and make more active use of all internal and external resources to increase investment in IoT technology. Besides, enterprises must improve knowledge stock and encourage employees to learn with a sound knowledge governance mechanism to continuously strengthen enterprises' absorptive capacity, innovation level, and innovation efficiency. Secondly, enterprises should strengthen the construction of IoT and form their own unique absorptive capacity that is difficult to imitate, to improve the search, understanding, transformation, and utilization of knowledge. Consequently, enterprises can lay a good foundation for the transformation of innovation mode and adaption to the dynamic environment, pressing innovation time and the instability of technological innovation.

However, due to objective limitations, the following problems are found. First, there is no in-depth analysis of the interactions between impact factors and open technological innovation, and different factors can be used to explore more new ways to improve the performance evaluation of enterprises' open technological innovation. Second, as for the development of the AI industry that affects the improvement of enterprises' open technological innovation capabilities, the mechanism of improving innovation performance through open technological innovation remains unclear. In the future, more in-depth discussions and analyses in these two aspects will be performed to continuously improve the open technological innovation evaluation model and more accurately evaluate enterprises' open technological innovation capabilities.

Different Levels of Smart and Sustainable Cities Construction Using e-Participation Tools in European and Central Asian Countries

An ontology-based security risk management model for information systems

Extending the study of finance-industry relationships

Subsidiary innovation performance: Balancing external knowledge sources and internal embeddedness

A framework of intellectual property protection strategies and open innovation

Managing intellectual property: Strategies for patent holders

Relationships between external knowledge, internal innovation, firms' open innovation performance, service innovation and business performance in the Pakistani hotel industry

Information Theory for Agents in Artificial Intelligence

Corporate Social Responsibility and Firm Performance in China's Manufacturing: A Global Perspective of Business Models

Upscaling evapotranspiration from the instantaneous to the daily time scale: Assessing six methods including an optimized coefficient based on worldwide eddy covariance flux network

Examining the Present and Future Integrated Role of Artificial Intelligence in the Business: A Survey Study on Corporate Sector

Industrial internet of things: Recent advances, enabling technologies and open challenges

Impact of sport on social cognition: an analysis based on structural equational models

Dynamically evaluating technological innovation efficiency of high-tech industry in China: Provincial, regional and industrial perspective

Green technology innovation and financial development: Do environmental regulation and innovation output matter?

Analysis of using blockchain to protect the privacy of drone big data

Fine-Grained Visual Computing Based on Deep Learning

6G-enabled network in box for internet of connected vehicles

Analysis of the security of internet of multimedia things

Vision-Based Measurement: Actualities and Developing Trends in Automated Container Terminals

The role of open innovation in developing an entrepreneurial support ecosystem

Reconceptualizing the interventions of open innovation systems between the nexus of quadruple organization cultural dynamics and performance

Reactive SPS for sol-gel alumina samples: Structure, sintering behavior, and mechanical properties

Mechanical characterization of sol-gel alumina-based ceramics with intragranular reinforcement of multiwalled carbon nanotubes

Predicting the citation counts of individual papers via a BP neural network

Synergy and policy-making in German innovation systems: Smart Specialisation Strategies at national, regional, local levels?

Innovation capability in SMEs: A systematic review of the literature

Ambidexterity and sustainable balanced performance in knowledge-based organisations: investigating the role of open innovation

Impact of Selected Intelligent Systems in Logistics on the Creation of a Sustainable Market Position of Manufacturing Companies

Analysis of the relationship between open innovation, knowledge management capability and dual innovation

Analogical reasoning, cognition, and the response to technological change: lessons from mobile communication

Rethinking Implicit Leadership Theories: Tomorrow's Leaders are Collective, Generative, and Adaptive

Research on the Relationship between Structural Hole Location, Knowledge Management and Cooperative Innovation Performance in Artificial Intelligence

On Scalability of Association-rulebased recommendation: A unified distributed-computing framework

High-Performance All-Polymer Solar Cells with a Pseudo-Bilayer Configuration Enabled by a Stepwise Optimization Strategy

The spatial-temporal variation and convergence of green innovation efficiency in the Yangtze River Economic Belt in China

Internal and External Coordinated Open Innovation Ecosystems: Concept Building and Applying to Shanghai Zizhu International Education Park

Spatial correlation analysis of low-carbon innovation: A case study of manufacturing patents in China

Open innovation in the public sector: creating public value through civic hackathons

Robustness of the active rotary inertia driver system for structural swing vibration control subjected to multitype hazard excitations

Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations

The authors declare that they have no conflicts of interest.