Identifying Feature Relevance Using a Random Forest. Rogers, J. & Gunn, S. Volume 3940, Saunders, C., Grobelnik, M., Gunn, S., & Shawe-Taylor, J., editors. Subspace, Latent Structure and Feature Selection, pages 173--184. Springer, Berlin / Heidelberg, 2006. doi abstract bibtex It is known that feature selection and feature relevance can benefit the performance and interpretation of machine learning algorithms. Here we consider feature selection within a Random Forest framework. A feature selection technique is introduced that combines hypothesis testing with an approximation to the expected performance of an irrelevant feature during Random Forest construction. It is demonstrated that the lack of implicit feature selection within Random Forest has an adverse effect on the accuracy and efficiency of the algorithm. It is also shown that irrelevant features can slow the rate of error convergence and a theoretical justification of this effect is given.
@inbook{Rogers:2006aa,
Abstract = {It is known that feature selection and feature relevance can benefit the performance and interpretation of machine learning algorithms. Here we consider feature selection within a Random Forest framework. A feature selection technique is introduced that combines hypothesis testing with an approximation to the expected performance of an irrelevant feature during Random Forest construction.
It is demonstrated that the lack of implicit feature selection within Random Forest has an adverse effect on the accuracy and efficiency of the algorithm. It is also shown that irrelevant features can slow the rate of error convergence and a theoretical justification of this effect is given.},
Address = {Berlin / Heidelberg},
Author = {Rogers, Jeremy and Gunn, Steve},
Chapter = {Identifying Feature Relevance Using a Random Forest},
Date-Added = {2008-05-10 13:39:38 -0400},
Date-Modified = {2009-04-16 15:14:55 -0400},
Doi = {10.1007/11752790_12},
Editor = {Saunders, C. and Grobelnik, M. and Gunn, S. and Shawe-Taylor, J.},
Keywords = {feature selection; descriptor importance; variable importance; bagging},
Pages = {173--184},
Publisher = {Springer},
Series = {Lecture Notes in Computer Science},
Timescited = {0},
Title = {Subspace, Latent Structure and Feature Selection},
Volume = {3940},
Year = {2006},
Bdsk-File-1 = {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}}
Downloads: 0
{"_id":"ZMC9Js74PTDpPG5mZ","bibbaseid":"rogers-gunn-subspacelatentstructureandfeatureselection-2006","downloads":0,"creationDate":"2016-02-18T13:03:32.832Z","title":"Subspace, Latent Structure and Feature Selection","author_short":["Rogers, J.","Gunn, S."],"year":2006,"bibtype":"inbook","biburl":"https://dl.dropboxusercontent.com/u/26998770/main.bib","bibdata":{"bibtype":"inbook","type":"inbook","abstract":"It is known that feature selection and feature relevance can benefit the performance and interpretation of machine learning algorithms. Here we consider feature selection within a Random Forest framework. A feature selection technique is introduced that combines hypothesis testing with an approximation to the expected performance of an irrelevant feature during Random Forest construction. It is demonstrated that the lack of implicit feature selection within Random Forest has an adverse effect on the accuracy and efficiency of the algorithm. It is also shown that irrelevant features can slow the rate of error convergence and a theoretical justification of this effect is given.","address":"Berlin / Heidelberg","author":[{"propositions":[],"lastnames":["Rogers"],"firstnames":["Jeremy"],"suffixes":[]},{"propositions":[],"lastnames":["Gunn"],"firstnames":["Steve"],"suffixes":[]}],"chapter":"Identifying Feature Relevance Using a Random Forest","date-added":"2008-05-10 13:39:38 -0400","date-modified":"2009-04-16 15:14:55 -0400","doi":"10.1007/11752790_12","editor":[{"propositions":[],"lastnames":["Saunders"],"firstnames":["C."],"suffixes":[]},{"propositions":[],"lastnames":["Grobelnik"],"firstnames":["M."],"suffixes":[]},{"propositions":[],"lastnames":["Gunn"],"firstnames":["S."],"suffixes":[]},{"propositions":[],"lastnames":["Shawe-Taylor"],"firstnames":["J."],"suffixes":[]}],"keywords":"feature selection; descriptor importance; variable importance; bagging","pages":"173--184","publisher":"Springer","series":"Lecture Notes in Computer Science","timescited":"0","title":"Subspace, Latent Structure and Feature Selection","volume":"3940","year":"2006","bdsk-file-1":"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","bibtex":"@inbook{Rogers:2006aa,\n\tAbstract = {It is known that feature selection and feature relevance can benefit the performance and interpretation of machine learning algorithms. Here we consider feature selection within a Random Forest framework. A feature selection technique is introduced that combines hypothesis testing with an approximation to the expected performance of an irrelevant feature during Random Forest construction.\n\nIt is demonstrated that the lack of implicit feature selection within Random Forest has an adverse effect on the accuracy and efficiency of the algorithm. It is also shown that irrelevant features can slow the rate of error convergence and a theoretical justification of this effect is given.},\n\tAddress = {Berlin / Heidelberg},\n\tAuthor = {Rogers, Jeremy and Gunn, Steve},\n\tChapter = {Identifying Feature Relevance Using a Random Forest},\n\tDate-Added = {2008-05-10 13:39:38 -0400},\n\tDate-Modified = {2009-04-16 15:14:55 -0400},\n\tDoi = {10.1007/11752790_12},\n\tEditor = {Saunders, C. and Grobelnik, M. and Gunn, S. and Shawe-Taylor, J.},\n\tKeywords = {feature selection; descriptor importance; variable importance; bagging},\n\tPages = {173--184},\n\tPublisher = {Springer},\n\tSeries = {Lecture Notes in Computer Science},\n\tTimescited = {0},\n\tTitle = {Subspace, Latent Structure and Feature Selection},\n\tVolume = {3940},\n\tYear = {2006},\n\tBdsk-File-1 = {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}}\n\n","author_short":["Rogers, J.","Gunn, S."],"editor_short":["Saunders, C.","Grobelnik, M.","Gunn, S.","Shawe-Taylor, J."],"key":"Rogers:2006aa","id":"Rogers:2006aa","bibbaseid":"rogers-gunn-subspacelatentstructureandfeatureselection-2006","role":"author","urls":{},"keyword":["feature selection; descriptor importance; variable importance; bagging"],"downloads":0},"search_terms":["subspace","latent","structure","feature","selection","rogers","gunn"],"keywords":["feature selection; descriptor importance; variable importance; bagging"],"authorIDs":[],"dataSources":["c5japf9eAQRaeMS4h"]}