Application-Independent Feature Construction Based onAlmost-Closedness Properties

Feature construction has been studied extensively, including for 0/1 data samples. Given the recent breakthroughs in closedness-related constraint-based mining, we are considering its impact on feature construction for classification tasks. We investigate the use of condensed representations of freq...

Full description

Saved in:
Bibliographic Details
Published in:Knowledge and information systems 2011-03, Vol.30, p.87-111
Main Authors: Gay, Dominique, Selmaoui-Folcher, Nazha, Boulicaut, Jean-François
Format: Article
Language:English
Subjects:
Computer Science
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 111
container_issue
container_start_page 87
container_title Knowledge and information systems
container_volume 30
creator Gay, Dominique
Selmaoui-Folcher, Nazha
Boulicaut, Jean-François
description Feature construction has been studied extensively, including for 0/1 data samples. Given the recent breakthroughs in closedness-related constraint-based mining, we are considering its impact on feature construction for classification tasks. We investigate the use of condensed representations of frequent itemsets based on closedness properties as new features. These itemset types have been proposed to avoid set counting in difficult association rule mining tasks, i.e., when datais noisy and/or highly correlated. However, our guess is that their intrinsic properties (say the maximality for the closed itemsets and the minimality for the delta-free itemsets) should have an impact on feature quality.Understanding this remains fairly open and we discuss these issues thanks to itemset properties on the one hand and an experimental validation on various data sets (possibly noisy) on the other hand.
format article
fullrecord <record><control><sourceid>hal</sourceid><recordid>TN_cdi_hal_primary_oai_HAL_hal_01354374v1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>oai_HAL_hal_01354374v1</sourcerecordid><originalsourceid>FETCH-hal_primary_oai_HAL_hal_01354374v13</originalsourceid><addsrcrecordid>eNqVissKwjAURIMoWB__kK2LQK-pDS5rsVRwoeC-BL1iJCYhNwr-vQr9ATczc4YzYFm-hLWQAOWw3yCVGrMJ0T3PQZUAGTtWIVhz1sl4J3buggG_4RJvUKdnRF57Ryk-zz-BbzThhXtX2YenJGrrv-yQiB-iDxiTQZqx0VVbwnnfU7Zotqe6FTdtuxDNQ8d357Xp2mrf_b4c5KqQqniB_Mf9AEj0RUo</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Application-Independent Feature Construction Based onAlmost-Closedness Properties</title><source>ABI/INFORM Global</source><source>Springer Link</source><creator>Gay, Dominique ; Selmaoui-Folcher, Nazha ; Boulicaut, Jean-François</creator><creatorcontrib>Gay, Dominique ; Selmaoui-Folcher, Nazha ; Boulicaut, Jean-François</creatorcontrib><description>Feature construction has been studied extensively, including for 0/1 data samples. Given the recent breakthroughs in closedness-related constraint-based mining, we are considering its impact on feature construction for classification tasks. We investigate the use of condensed representations of frequent itemsets based on closedness properties as new features. These itemset types have been proposed to avoid set counting in difficult association rule mining tasks, i.e., when datais noisy and/or highly correlated. However, our guess is that their intrinsic properties (say the maximality for the closed itemsets and the minimality for the delta-free itemsets) should have an impact on feature quality.Understanding this remains fairly open and we discuss these issues thanks to itemset properties on the one hand and an experimental validation on various data sets (possibly noisy) on the other hand.</description><identifier>ISSN: 0219-1377</identifier><identifier>EISSN: 0219-3116</identifier><language>eng</language><publisher>Springer</publisher><subject>Computer Science</subject><ispartof>Knowledge and information systems, 2011-03, Vol.30, p.87-111</ispartof><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><orcidid>0000-0002-0671-4616 ; 0000-0003-1667-3819 ; 0000-0002-0671-4616 ; 0000-0003-1667-3819</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,776,780,881</link.rule.ids><backlink>$$Uhttps://hal.science/hal-01354374$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Gay, Dominique</creatorcontrib><creatorcontrib>Selmaoui-Folcher, Nazha</creatorcontrib><creatorcontrib>Boulicaut, Jean-François</creatorcontrib><title>Application-Independent Feature Construction Based onAlmost-Closedness Properties</title><title>Knowledge and information systems</title><description>Feature construction has been studied extensively, including for 0/1 data samples. Given the recent breakthroughs in closedness-related constraint-based mining, we are considering its impact on feature construction for classification tasks. We investigate the use of condensed representations of frequent itemsets based on closedness properties as new features. These itemset types have been proposed to avoid set counting in difficult association rule mining tasks, i.e., when datais noisy and/or highly correlated. However, our guess is that their intrinsic properties (say the maximality for the closed itemsets and the minimality for the delta-free itemsets) should have an impact on feature quality.Understanding this remains fairly open and we discuss these issues thanks to itemset properties on the one hand and an experimental validation on various data sets (possibly noisy) on the other hand.</description><subject>Computer Science</subject><issn>0219-1377</issn><issn>0219-3116</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2011</creationdate><recordtype>article</recordtype><recordid>eNqVissKwjAURIMoWB__kK2LQK-pDS5rsVRwoeC-BL1iJCYhNwr-vQr9ATczc4YzYFm-hLWQAOWw3yCVGrMJ0T3PQZUAGTtWIVhz1sl4J3buggG_4RJvUKdnRF57Ryk-zz-BbzThhXtX2YenJGrrv-yQiB-iDxiTQZqx0VVbwnnfU7Zotqe6FTdtuxDNQ8d357Xp2mrf_b4c5KqQqniB_Mf9AEj0RUo</recordid><startdate>201103</startdate><enddate>201103</enddate><creator>Gay, Dominique</creator><creator>Selmaoui-Folcher, Nazha</creator><creator>Boulicaut, Jean-François</creator><general>Springer</general><scope>1XC</scope><orcidid>https://orcid.org/0000-0002-0671-4616</orcidid><orcidid>https://orcid.org/0000-0003-1667-3819</orcidid><orcidid>https://orcid.org/0000-0002-0671-4616</orcidid><orcidid>https://orcid.org/0000-0003-1667-3819</orcidid></search><sort><creationdate>201103</creationdate><title>Application-Independent Feature Construction Based onAlmost-Closedness Properties</title><author>Gay, Dominique ; Selmaoui-Folcher, Nazha ; Boulicaut, Jean-François</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-hal_primary_oai_HAL_hal_01354374v13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2011</creationdate><topic>Computer Science</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Gay, Dominique</creatorcontrib><creatorcontrib>Selmaoui-Folcher, Nazha</creatorcontrib><creatorcontrib>Boulicaut, Jean-François</creatorcontrib><collection>Hyper Article en Ligne (HAL)</collection><jtitle>Knowledge and information systems</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Gay, Dominique</au><au>Selmaoui-Folcher, Nazha</au><au>Boulicaut, Jean-François</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Application-Independent Feature Construction Based onAlmost-Closedness Properties</atitle><jtitle>Knowledge and information systems</jtitle><date>2011-03</date><risdate>2011</risdate><volume>30</volume><spage>87</spage><epage>111</epage><pages>87-111</pages><issn>0219-1377</issn><eissn>0219-3116</eissn><abstract>Feature construction has been studied extensively, including for 0/1 data samples. Given the recent breakthroughs in closedness-related constraint-based mining, we are considering its impact on feature construction for classification tasks. We investigate the use of condensed representations of frequent itemsets based on closedness properties as new features. These itemset types have been proposed to avoid set counting in difficult association rule mining tasks, i.e., when datais noisy and/or highly correlated. However, our guess is that their intrinsic properties (say the maximality for the closed itemsets and the minimality for the delta-free itemsets) should have an impact on feature quality.Understanding this remains fairly open and we discuss these issues thanks to itemset properties on the one hand and an experimental validation on various data sets (possibly noisy) on the other hand.</abstract><pub>Springer</pub><orcidid>https://orcid.org/0000-0002-0671-4616</orcidid><orcidid>https://orcid.org/0000-0003-1667-3819</orcidid><orcidid>https://orcid.org/0000-0002-0671-4616</orcidid><orcidid>https://orcid.org/0000-0003-1667-3819</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 0219-1377
ispartof Knowledge and information systems, 2011-03, Vol.30, p.87-111
issn 0219-1377
0219-3116
language eng
recordid cdi_hal_primary_oai_HAL_hal_01354374v1
source ABI/INFORM Global; Springer Link
subjects Computer Science
title Application-Independent Feature Construction Based onAlmost-Closedness Properties
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-13T20%3A47%3A56IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-hal&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Application-Independent%20Feature%20Construction%20Based%20onAlmost-Closedness%20Properties&rft.jtitle=Knowledge%20and%20information%20systems&rft.au=Gay,%20Dominique&rft.date=2011-03&rft.volume=30&rft.spage=87&rft.epage=111&rft.pages=87-111&rft.issn=0219-1377&rft.eissn=0219-3116&rft_id=info:doi/&rft_dat=%3Chal%3Eoai_HAL_hal_01354374v1%3C/hal%3E%3Cgrp_id%3Ecdi_FETCH-hal_primary_oai_HAL_hal_01354374v13%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true