Show simple item record

dc.contributor.authorLoughrey, John
dc.contributor.authorCunningham, Padraig
dc.date.accessioned2008-01-24T11:14:17Z
dc.date.available2008-01-24T11:14:17Z
dc.date.issued2005-05-11
dc.identifier.citationLoughrey, John; Cunningham, Padraig. 'Using Early-Stopping to Avoid Overfitting in Wrapper-Based Feature Selection Employing Stochastic Search'. - Dublin, Trinity College Dublin, Department of Computer Science, TCD-CS-2005-37, 2005, pp6en
dc.identifier.otherTCD-CS-2005-37
dc.identifier.urihttp://hdl.handle.net/2262/13359
dc.description.abstractIt is acknowledged that overfitting can occur in feature selection using the wrapper method when there is a limited amount of training data available. It has also been shown that the severity of overfitting is related to the intensity of the search algorithm used during this process. In this paper we show that two stochastic search techniques (Simulated Annealing and Genetic Algorithms) that can be used for wrapper-based feature selection are susceptible to overfitting in this way. However, because of their stochastic nature, these algorithms can be stopped early to prevent overfitting. We present a framework that implements early-stopping for both of these stochastic search techniques and we show that this is successful in reducing the effects of overfitting and in increasing generalisation accuracy in most cases.en
dc.format.extent179908 bytes
dc.format.mimetypeapplication/pdf
dc.language.isoenen
dc.publisherTrinity College Dublin, Department of Computer Scienceen
dc.relation.ispartofseriesComputer Science Technical Reporten
dc.relation.ispartofseriesTCD-CS-2005-37en
dc.relation.haspartTCD-CS-[no.]en
dc.subjectComputer Scienceen
dc.titleUsing Early-Stopping to Avoid Overfitting in Wrapper-Based Feature Selection Employing Stochastic Searchen
dc.typeTechnical Reporten
dc.identifier.rssurihttps://www.cs.tcd.ie/publications/tech-reports/reports.05/TCD-CS-2005-37.pdf


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record