Review and Evaluation of Feature Selection Algorithms in Synthetic Problems
作者： L. A. Belanche , F. F. González
DOI:
关键词:
摘要: Abstract: The main purpose of Feature Subset Selection is to find a reduced subsetof attributes from data set described by feature set. task featureselectionalgorithm (FSA) provide with computational solution motivated certaindefinition relevance or reliable evaluation measure. In this paper several funda-mental algorithms are studied assess their performance in controlled experimentalscenario. A measure evaluate FSAs devised that computes the degree matchingbetween output given FSA and known optimal solutions. An extensiveexperimental study on synthetic problems carried out behaviour ofthe terms accuracy size as function relevance,irrelevance, redundancy samples. experimentalconditions facilitate derivation better-supported meaningful conclusions.Keywords:Feature Algorithms; Empirical Evaluations; Attribute relevanceand redundancy.1 INTRODUCTION
arxiv.org
harvard.edu
128.84.21.199参考文章(23)
Thomas G. Dietterich, Hussein Almuallim, Learning with many irrelevant features national conference on artificial intelligence. pp. 547- 552 ,(1991)
Patrick H. Winston, Learning Structural Descriptions From Examples The Psychology of Computer Vision. ,(1970)
Manoranjan Dash, Huan Liu, Hiroshi Motoda, Consistency Based Feature Selection pacific asia conference on knowledge discovery and data mining. pp. 98- 109 ,(2000) , 10.1007/3-540-45571-X_12
M. Hall, Correlation-based Feature Selection for Machine Learning PhD Thesis, Waikato Univer-sity. ,(1998)
Huan Liu, Rudy Setiono, Incremental Feature Selection Applied Intelligence. ,vol. 9, pp. 217- 230 ,(1998) , 10.1023/A:1008363719778
Jack Sklansky, Mineichi Kudo, A COMPARATIVE EVALUATION OF MEDIUM-AND LARGE-SCALE FEATURE SELECTORS FOR PATTERN CLASSIFIERS Kybernetika. ,vol. 34, pp. 429- 434 ,(1998)
Huan Liul, Hiroshi Motoda, Manoranjan Dash, A Monotonic Measure for Optimal Feature Selection european conference on machine learning. pp. 101- 106 ,(1998) , 10.1007/BFB0026678
Narendra, Fukunaga, A Branch and Bound Algorithm for Feature Subset Selection IEEE Transactions on Computers. ,vol. 26, pp. 917- 922 ,(1977) , 10.1109/TC.1977.1674939
Manoranjan Dash, Huan Liu, Hybrid Search of Feature Subsets pacific rim international conference on artificial intelligence. pp. 238- 249 ,(1998) , 10.1007/BFB0095273
Rich Caruana, Dayne Freitag, Greedy Attribute Selection Machine Learning Proceedings 1994. pp. 28- 36 ,(1994) , 10.1016/B978-1-55860-335-6.50012-X