Artykuł w czasopiśmie
Brak miniatury
Licencja
ClosedAccessDostęp zamknięty
 

Scalable Machine Learning with Granulated Data Summaries: A Case of Feature Selection

Uproszczony widok
dc.abstract.enWe investigate how to use the histogram-based data summaries that are created and stored by one of the approximate database engines available in the market, for the purposes of redesigning and accelerating machine learning algorithms. As an example, we consider one of popular minimum redundancy maximum relevance (mRMR) feature selection methods based on mutual information. We use granulated data summaries to approximately calculate the entropy-based mutual information scores and observe the mRMR results compared to the case of working with the actual scores derived from the original data.
dc.affiliationUniwersytet Warszawski
dc.conference.countryPolska
dc.conference.datefinish2017-06-29
dc.conference.datestart2017-06-26
dc.conference.placezAKOPANE
dc.conference.seriesInternational Symposium on Foundations of Intelligent Systems
dc.conference.seriesInternational Symposium on Foundations of Intelligent Systems
dc.conference.seriesshortcutISMIS
dc.conference.shortcutISMIS 2017
dc.conference.weblinkhttp://ismis2017.ii.pw.edu.pl/index.php
dc.contributor.authorŚlęzak, Dominik
dc.contributor.authorBetliński, Paweł
dc.contributor.authorChądzyńska-Krasowska, Agnieszka
dc.date.accessioned2024-01-25T19:43:30Z
dc.date.available2024-01-25T19:43:30Z
dc.date.issued2017
dc.description.financeNie dotyczy
dc.identifier.doi10.1007/978-3-319-60438-1_51
dc.identifier.urihttps://repozytorium.uw.edu.pl//handle/item/119039
dc.identifier.weblinkhttps://link.springer.com/chapter/10.1007%2F978-3-319-60438-1_51
dc.languageeng
dc.pbn.affiliationcomputer and information sciences
dc.relation.pages519-529
dc.rightsClosedAccess
dc.sciencecloudnosend
dc.subject.enData granulation
dc.subject.enApproximate query
dc.subject.enFeature selection
dc.titleScalable Machine Learning with Granulated Data Summaries: A Case of Feature Selection
dc.typeJournalArticle
dspace.entity.typePublication