Show simple item record  

dc.contributor.authorHall, Mark A.
dc.contributor.authorSmith, Lloyd A.
dc.coverage.spatialConference held at Kasabov, Nikola, Robert Kozma, Kitty Ko, Robert O'Shea, George Coghill, Tom Gedeonen_NZ
dc.date.accessioned2008-12-02T01:28:52Z
dc.date.available2008-12-02T01:28:52Z
dc.date.issued1997
dc.identifier.citationHall, M. A. & Smith, L. A. (1997). Feature subset selection: a correlation based filter approach. In 1997 International Conference on Neural Information Processing and Intelligent Information Systems (pp. 855-858). Berlin: Springer.en_US
dc.identifier.urihttps://hdl.handle.net/10289/1515
dc.description.abstractRecent work has shown that feature subset selection can have a position affect on the performance of machine learning algorithms. Some algorithms can be slowed or their performance adversely affected by too much data some of which may be irrelevant or redundant to the learning task. Feature subset selection, then, is a method of enhancing the performance of learning algorithms, reducing the hypothesis search space, and, in some cases, reducing the storage requirement. This paper describes a feature subset selector that uses a correlation based heuristic to determine the goodness of feature subsets, and evaluates its effectiveness with three common ML algorithms: a decision tree inducer (C4.5), a naive Bayes classifier, and an instance based learner(IBI). Experiments using a number of standard data sets drawn from real and artificial domains are presented. Feature subset selection gave significant improvement for all three algorithms; C4.5 generated smaller decision trees.en_US
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherSpringeren_US
dc.rightsThis is an author’s version of an article published in 1997 International Conference on Neural Information Processing and Intelligent Information Systems. © Springer.en_US
dc.subjectcomputer scienceen_US
dc.subjectfeature selectionen_US
dc.subjectdecision treeen_US
dc.subjectnaive Bayesen_US
dc.titleFeature subset selection: a correlation based filter approachen_US
dc.typeConference Contributionen_US
dc.relation.isPartOf1997 International Conference on Neural Information Processing and Intelligent Information Systemsen_NZ
pubs.begin-page855en_NZ
pubs.elements-id23956
pubs.end-page858en_NZ
pubs.volumeProgress in Connectionist-based Information Systems. Volume 2en_NZ


Files in this item

This item appears in the following Collection(s)

Show simple item record