id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_zzyeey6arvaxbhnnhyc5xqfngy	Ahmed Al-Ani	A dependency-based search strategy for feature selection	2009	24	.pdf	application/pdf	5984	1188	78	aims at finding optimal feature subsets that can achieve better generalization Key words: Feature selection, search strategy, dependency, mutual and a search strategy to generate candidate feature subsets (1). filter method uses a measure that is independent of the predetermined classification algorithm to estimate the goodness of candidate subsets. Searching for the optimal subset, which can achieve the best performance methods attempt to estimate the relevance of subsets of features as the basis methods, the proposed search strategy utilizes dependency between feature The following methods were implemented to select feature subsets: genetic algorithms, particle swarm optimization, differential evolution selection, as the optimal subset of features is already known. The results obtained for the selection of subsets of 30 features DSS also outperformed the other three methods when selecting 30 features. Results for the selection of subsets of 30 features when using a k-NN H. John, "Wrappers for feature subset selection," Artificial Intelligence, special issue on relevance, vol.	./cache/work_zzyeey6arvaxbhnnhyc5xqfngy.pdf	./txt/work_zzyeey6arvaxbhnnhyc5xqfngy.txt