Publication:

Powershap: A Power-Full Shapley Feature Selection Method

Date

2023
Proceedings Paper
https://doi.org/10.1007/978-3-031-26387-3_5
Simple item page Full metadata Statistics
 
dc.contributor.authorVerhaeghe, Jarne
dc.contributor.authorVan Der Donckt, Jeroen
dc.contributor.authorOngenae, Femke
dc.contributor.authorVan Hoecke, Sofie
dc.contributor.imecauthorVerhaeghe, Jarne
dc.contributor.imecauthorVan Der Donckt, Jeroen
dc.contributor.imecauthorOngenae, Femke
dc.contributor.imecauthorVan Hoecke, Sofie
dc.contributor.orcidimecVerhaeghe, Jarne::0000-0002-3322-150X
dc.contributor.orcidimecOngenae, Femke::0000-0003-2529-5477
dc.contributor.orcidimecVan Hoecke, Sofie::0000-0002-7865-6793
dc.contributor.orcidimecVan Der Donckt, Jeroen::0000-0002-9620-888X
dc.date.accessioned2023-07-20T11:20:18Z
dc.date.available2023-06-17T20:17:25Z
dc.date.available2023-06-19T10:09:47Z
dc.date.available2023-07-20T11:20:18Z
dc.date.embargo2023-03-17
dc.date.issued2023
dc.description.wosFundingTextJarne Verhaeghe is funded by the Research Foundation Flanders (FWO, Ref. 1S59522N) and designed powershap. Jeroen Van Der Donckt implemented powershap as an sklearn component. Sofie Van Hoecke and Femke Ongenae supervised the project. A special thanks goes to Gilles Vandewiele for proof-reading the manuscript.
dc.identifier.doi10.1007/978-3-031-26387-3_5
dc.identifier.eisbn978-3-031-26387-3
dc.identifier.isbn978-3-031-26386-6
dc.identifier.issn2945-9133
dc.identifier.urihttps://imec-publications.be/handle/20.500.12860/41750
dc.publisherSPRINGER INTERNATIONAL PUBLISHING AG
dc.source.beginpage71
dc.source.conferenceEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD)
dc.source.conferencedateSEP 19-23, 2022
dc.source.conferencelocationGrenoble
dc.source.endpage87
dc.source.journalECML PKDD 2022: Machine Learning and Knowledge Discovery in Databases
dc.source.numberofpages17
dc.source.volume13713
dc.title

Powershap: A Power-Full Shapley Feature Selection Method

dc.typeProceedings paper
dspace.entity.typePublication
Files

Original bundle

Name:
978-3-031-26387-3_5.pdf
Size:
654.48 KB
Format:
Adobe Portable Document Format
Description:
Published version
Download
Publication available in collections:
Conference contributions