An experiment with association rules and classification: post-bagging and conviction
In this paper we study a new technique we call post-bagging, which consists in resampling parts of a classification model rather then the data. We do this with a particular kind of model: large sets of classification association rules, and in combination with ordinary best rule and weighted voting a...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 149 |
|---|---|
| container_issue | |
| container_start_page | 137 |
| container_title | |
| container_volume | 3735 |
| creator | Jorge, Alípio M. Azevedo, Paulo J. |
| description | In this paper we study a new technique we call post-bagging, which consists in resampling parts of a classification model rather then the data. We do this with a particular kind of model: large sets of classification association rules, and in combination with ordinary best rule and weighted voting approaches. We empirically evaluate the effects of the technique in terms of classification accuracy. We also discuss the predictive power of different metrics used for association rule mining, such as confidence, lift, conviction and X². We conclude that, for the described experimental conditions, post-bagging improves classification results and that the best metric is conviction.
Programa de Financiamento Plurianual de Unidades de I & D.
Comunidade Europeia (CE). Fundo Europeu de Desenvolvimento Regional (FEDER).
Fundação para a Ciência e a Tecnologia (FCT) - POSI/SRI/39630/2001/Class Project. |
| doi_str_mv | 10.1007/11563983_13 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>pascalfrancis_sprin</sourceid><recordid>TN_cdi_pascalfrancis_primary_17413620</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>17413620</sourcerecordid><originalsourceid>FETCH-LOGICAL-c322t-840c0a37d5b2537125a103633c7fff5b3bac551fff9757d03bb9497fa415224f3</originalsourceid><addsrcrecordid>eNpNkD1PwzAQhs2XRCmd-AFkYWAI-Hx2HLNVFV9SJZYyW44bB0NwojgU-Pe4lIFb7nTPo5PuJeQM6BVQKq8BRIGqRA24R2ZKlig4RShUKfbJBAqAHJGrA3KyBUwxpMUhmVCkLFeS4zGZxfhKUyFIhmJCVvOQ1V99Pfj3OozZpx9fMhNjZ70ZfRey4aOtY2bCOrNt2nvn7S-4yfoujnllmsaHZid0YePtFp6SI2faWM_--pQ8392uFg_58un-cTFf5hYZG_OSU0sNyrWomEAJTBigWCBa6ZwTFVbGCgFpVlLINcWqUlxJZzgIxrjDKbnY3e1NtKZ1gwnWR92nZ8zwrUFywILR5F3uvJhQaOpBV133FjVQvY1V_4s1uec7d7DG9HqoNz6OJrklY5ozJfAHwkdu3A</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>An experiment with association rules and classification: post-bagging and conviction</title><source>Springer Books</source><creator>Jorge, Alípio M. ; Azevedo, Paulo J.</creator><contributor>Hoffmann, Achim ; Motoda, Hiroshi ; Scheffer, Tobias</contributor><creatorcontrib>Jorge, Alípio M. ; Azevedo, Paulo J. ; Hoffmann, Achim ; Motoda, Hiroshi ; Scheffer, Tobias</creatorcontrib><description>In this paper we study a new technique we call post-bagging, which consists in resampling parts of a classification model rather then the data. We do this with a particular kind of model: large sets of classification association rules, and in combination with ordinary best rule and weighted voting approaches. We empirically evaluate the effects of the technique in terms of classification accuracy. We also discuss the predictive power of different metrics used for association rule mining, such as confidence, lift, conviction and X². We conclude that, for the described experimental conditions, post-bagging improves classification results and that the best metric is conviction.
Programa de Financiamento Plurianual de Unidades de I & D.
Comunidade Europeia (CE). Fundo Europeu de Desenvolvimento Regional (FEDER).
Fundação para a Ciência e a Tecnologia (FCT) - POSI/SRI/39630/2001/Class Project.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 3540292306</identifier><identifier>ISBN: 9783540292302</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 9783540316985</identifier><identifier>EISBN: 3540316981</identifier><identifier>DOI: 10.1007/11563983_13</identifier><language>eng</language><publisher>Berlin, Heidelberg: Springer Verlag</publisher><subject>Applied sciences ; Artificial intelligence ; Association Rule ; Association Rule Mining ; Association rules ; Classification ; Computer science; control theory; systems ; Decision Tree Inducer ; Exact sciences and technology ; Frequent Itemset ; Frequent Pattern Mining ; Science & Technology</subject><ispartof>Discovery Science, 2005, Vol.3735, p.137-149</ispartof><rights>Springer-Verlag Berlin Heidelberg 2005</rights><rights>2006 INIST-CNRS</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c322t-840c0a37d5b2537125a103633c7fff5b3bac551fff9757d03bb9497fa415224f3</citedby></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/11563983_13$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/11563983_13$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>309,310,775,776,780,785,786,789,4036,4037,27902,38232,41418,42487</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=17413620$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><contributor>Hoffmann, Achim</contributor><contributor>Motoda, Hiroshi</contributor><contributor>Scheffer, Tobias</contributor><creatorcontrib>Jorge, Alípio M.</creatorcontrib><creatorcontrib>Azevedo, Paulo J.</creatorcontrib><title>An experiment with association rules and classification: post-bagging and conviction</title><title>Discovery Science</title><description>In this paper we study a new technique we call post-bagging, which consists in resampling parts of a classification model rather then the data. We do this with a particular kind of model: large sets of classification association rules, and in combination with ordinary best rule and weighted voting approaches. We empirically evaluate the effects of the technique in terms of classification accuracy. We also discuss the predictive power of different metrics used for association rule mining, such as confidence, lift, conviction and X². We conclude that, for the described experimental conditions, post-bagging improves classification results and that the best metric is conviction.
Programa de Financiamento Plurianual de Unidades de I & D.
Comunidade Europeia (CE). Fundo Europeu de Desenvolvimento Regional (FEDER).
Fundação para a Ciência e a Tecnologia (FCT) - POSI/SRI/39630/2001/Class Project.</description><subject>Applied sciences</subject><subject>Artificial intelligence</subject><subject>Association Rule</subject><subject>Association Rule Mining</subject><subject>Association rules</subject><subject>Classification</subject><subject>Computer science; control theory; systems</subject><subject>Decision Tree Inducer</subject><subject>Exact sciences and technology</subject><subject>Frequent Itemset</subject><subject>Frequent Pattern Mining</subject><subject>Science & Technology</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>3540292306</isbn><isbn>9783540292302</isbn><isbn>9783540316985</isbn><isbn>3540316981</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2005</creationdate><recordtype>conference_proceeding</recordtype><recordid>eNpNkD1PwzAQhs2XRCmd-AFkYWAI-Hx2HLNVFV9SJZYyW44bB0NwojgU-Pe4lIFb7nTPo5PuJeQM6BVQKq8BRIGqRA24R2ZKlig4RShUKfbJBAqAHJGrA3KyBUwxpMUhmVCkLFeS4zGZxfhKUyFIhmJCVvOQ1V99Pfj3OozZpx9fMhNjZ70ZfRey4aOtY2bCOrNt2nvn7S-4yfoujnllmsaHZid0YePtFp6SI2faWM_--pQ8392uFg_58un-cTFf5hYZG_OSU0sNyrWomEAJTBigWCBa6ZwTFVbGCgFpVlLINcWqUlxJZzgIxrjDKbnY3e1NtKZ1gwnWR92nZ8zwrUFywILR5F3uvJhQaOpBV133FjVQvY1V_4s1uec7d7DG9HqoNz6OJrklY5ozJfAHwkdu3A</recordid><startdate>2005</startdate><enddate>2005</enddate><creator>Jorge, Alípio M.</creator><creator>Azevedo, Paulo J.</creator><general>Springer Verlag</general><general>Springer Berlin Heidelberg</general><general>Springer</general><scope>RCLKO</scope><scope>IQODW</scope></search><sort><creationdate>2005</creationdate><title>An experiment with association rules and classification: post-bagging and conviction</title><author>Jorge, Alípio M. ; Azevedo, Paulo J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c322t-840c0a37d5b2537125a103633c7fff5b3bac551fff9757d03bb9497fa415224f3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2005</creationdate><topic>Applied sciences</topic><topic>Artificial intelligence</topic><topic>Association Rule</topic><topic>Association Rule Mining</topic><topic>Association rules</topic><topic>Classification</topic><topic>Computer science; control theory; systems</topic><topic>Decision Tree Inducer</topic><topic>Exact sciences and technology</topic><topic>Frequent Itemset</topic><topic>Frequent Pattern Mining</topic><topic>Science & Technology</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Jorge, Alípio M.</creatorcontrib><creatorcontrib>Azevedo, Paulo J.</creatorcontrib><collection>RCAAP open access repository</collection><collection>Pascal-Francis</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Jorge, Alípio M.</au><au>Azevedo, Paulo J.</au><au>Hoffmann, Achim</au><au>Motoda, Hiroshi</au><au>Scheffer, Tobias</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>An experiment with association rules and classification: post-bagging and conviction</atitle><btitle>Discovery Science</btitle><date>2005</date><risdate>2005</risdate><volume>3735</volume><spage>137</spage><epage>149</epage><pages>137-149</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>3540292306</isbn><isbn>9783540292302</isbn><eisbn>9783540316985</eisbn><eisbn>3540316981</eisbn><abstract>In this paper we study a new technique we call post-bagging, which consists in resampling parts of a classification model rather then the data. We do this with a particular kind of model: large sets of classification association rules, and in combination with ordinary best rule and weighted voting approaches. We empirically evaluate the effects of the technique in terms of classification accuracy. We also discuss the predictive power of different metrics used for association rule mining, such as confidence, lift, conviction and X². We conclude that, for the described experimental conditions, post-bagging improves classification results and that the best metric is conviction.
Programa de Financiamento Plurianual de Unidades de I & D.
Comunidade Europeia (CE). Fundo Europeu de Desenvolvimento Regional (FEDER).
Fundação para a Ciência e a Tecnologia (FCT) - POSI/SRI/39630/2001/Class Project.</abstract><cop>Berlin, Heidelberg</cop><pub>Springer Verlag</pub><doi>10.1007/11563983_13</doi><tpages>13</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0302-9743 |
| ispartof | Discovery Science, 2005, Vol.3735, p.137-149 |
| issn | 0302-9743 1611-3349 |
| language | eng |
| recordid | cdi_pascalfrancis_primary_17413620 |
| source | Springer Books |
| subjects | Applied sciences Artificial intelligence Association Rule Association Rule Mining Association rules Classification Computer science control theory systems Decision Tree Inducer Exact sciences and technology Frequent Itemset Frequent Pattern Mining Science & Technology |
| title | An experiment with association rules and classification: post-bagging and conviction |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-29T22%3A36%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-pascalfrancis_sprin&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=An%20experiment%20with%20association%20rules%20and%20classification:%20post-bagging%20and%20conviction&rft.btitle=Discovery%20Science&rft.au=Jorge,%20Al%C3%ADpio%20M.&rft.date=2005&rft.volume=3735&rft.spage=137&rft.epage=149&rft.pages=137-149&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=3540292306&rft.isbn_list=9783540292302&rft_id=info:doi/10.1007/11563983_13&rft_dat=%3Cpascalfrancis_sprin%3E17413620%3C/pascalfrancis_sprin%3E%3Curl%3E%3C/url%3E&rft.eisbn=9783540316985&rft.eisbn_list=3540316981&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |