Partial imputation to improve predictive modelling in insurance risk classification using a hybrid positive selection algorithm and correlation-based feature selection
We propose a hybrid missing data imputation technique using positive selection and correlation-based feature selection for insurance data. The hybrid is used to help supervised learning methods improve their classification accuracy and resilience in the presence of increasing missing data. The posit...
Gespeichert in:
| Veröffentlicht in: | Current science (Bangalore) 2012-09, Vol.103 (6), p.697-705 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 705 |
|---|---|
| container_issue | 6 |
| container_start_page | 697 |
| container_title | Current science (Bangalore) |
| container_volume | 103 |
| creator | Duma, Mlungisi Twala, Bhekisipho Nelwamondo, Fulufhelo V. Marwala, Tshilidzi |
| description | We propose a hybrid missing data imputation technique using positive selection and correlation-based feature selection for insurance data. The hybrid is used to help supervised learning methods improve their classification accuracy and resilience in the presence of increasing missing data. The positive selection algorithm searches for potential candidates for imputation and the correlation-based feature selection method searches for attributes have a significant effect on the target outcome. The imputation is performed only on those attributes that have an impact on the target outcome. The results show that the classification accuracy and resilience of supervised learning methods improve significantly when applied with the imputation strategy under these assumptions. |
| format | Article |
| fullrecord | <record><control><sourceid>jstor_proqu</sourceid><recordid>TN_cdi_proquest_miscellaneous_1221890791</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><jstor_id>24088803</jstor_id><sourcerecordid>24088803</sourcerecordid><originalsourceid>FETCH-LOGICAL-j210t-3e45f1680c2a6ebc768781b521ab2d36236c601d93acdbe5416bb27c3582c0843</originalsourceid><addsrcrecordid>eNpNj8tqwzAQRb1ooWnaTyho2Y1BDz_kZQl9QSBdtGszksaJUtlyJbmQL-pv1km6KAzMDJx74F5kC0oZy4Vs2FV2HeOeUi44bRbZzxuEZMER249TgmT9QJI_fsF_IxkDGquTnc_eG3TODltih3niFGDQSIKNn0Q7iNF2Vp8FUzxiQHYHFawho4_2pIjoUJ8IcFsfbNr1BAZDtA8B3SmbK4hoSIeQpvAvcZNdduAi3v7tZfbx9Pi-esnXm-fX1cM633NGUy6wKDtWSao5VKh0XclaMlVyBoobUXFR6Yoy0wjQRmFZsEopXmtRSq6pLMQyuz975_5fE8bU9jbquTgM6KfYMs6ZbGjdsBm9O6P7mHxox2B7CIeWF1RKSYX4BU-wd-Y</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1221890791</pqid></control><display><type>article</type><title>Partial imputation to improve predictive modelling in insurance risk classification using a hybrid positive selection algorithm and correlation-based feature selection</title><source>Jstor Complete Legacy</source><source>EZB Electronic Journals Library</source><creator>Duma, Mlungisi ; Twala, Bhekisipho ; Nelwamondo, Fulufhelo V. ; Marwala, Tshilidzi</creator><creatorcontrib>Duma, Mlungisi ; Twala, Bhekisipho ; Nelwamondo, Fulufhelo V. ; Marwala, Tshilidzi</creatorcontrib><description>We propose a hybrid missing data imputation technique using positive selection and correlation-based feature selection for insurance data. The hybrid is used to help supervised learning methods improve their classification accuracy and resilience in the presence of increasing missing data. The positive selection algorithm searches for potential candidates for imputation and the correlation-based feature selection method searches for attributes have a significant effect on the target outcome. The imputation is performed only on those attributes that have an impact on the target outcome. The results show that the classification accuracy and resilience of supervised learning methods improve significantly when applied with the imputation strategy under these assumptions.</description><identifier>ISSN: 0011-3891</identifier><language>eng</language><publisher>Current Science Association</publisher><subject>Credit risk ; Data imputation ; Datasets ; Information classification ; Insurance risk ; Machine learning ; Missing data ; Positive selection ; Predictive modeling ; RESEARCH COMMUNICATIONS ; Sensors</subject><ispartof>Current science (Bangalore), 2012-09, Vol.103 (6), p.697-705</ispartof><rights>2012 Current Science Association</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.jstor.org/stable/pdf/24088803$$EPDF$$P50$$Gjstor$$H</linktopdf><linktohtml>$$Uhttps://www.jstor.org/stable/24088803$$EHTML$$P50$$Gjstor$$H</linktohtml><link.rule.ids>314,780,784,803,58017,58250</link.rule.ids></links><search><creatorcontrib>Duma, Mlungisi</creatorcontrib><creatorcontrib>Twala, Bhekisipho</creatorcontrib><creatorcontrib>Nelwamondo, Fulufhelo V.</creatorcontrib><creatorcontrib>Marwala, Tshilidzi</creatorcontrib><title>Partial imputation to improve predictive modelling in insurance risk classification using a hybrid positive selection algorithm and correlation-based feature selection</title><title>Current science (Bangalore)</title><description>We propose a hybrid missing data imputation technique using positive selection and correlation-based feature selection for insurance data. The hybrid is used to help supervised learning methods improve their classification accuracy and resilience in the presence of increasing missing data. The positive selection algorithm searches for potential candidates for imputation and the correlation-based feature selection method searches for attributes have a significant effect on the target outcome. The imputation is performed only on those attributes that have an impact on the target outcome. The results show that the classification accuracy and resilience of supervised learning methods improve significantly when applied with the imputation strategy under these assumptions.</description><subject>Credit risk</subject><subject>Data imputation</subject><subject>Datasets</subject><subject>Information classification</subject><subject>Insurance risk</subject><subject>Machine learning</subject><subject>Missing data</subject><subject>Positive selection</subject><subject>Predictive modeling</subject><subject>RESEARCH COMMUNICATIONS</subject><subject>Sensors</subject><issn>0011-3891</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><recordid>eNpNj8tqwzAQRb1ooWnaTyho2Y1BDz_kZQl9QSBdtGszksaJUtlyJbmQL-pv1km6KAzMDJx74F5kC0oZy4Vs2FV2HeOeUi44bRbZzxuEZMER249TgmT9QJI_fsF_IxkDGquTnc_eG3TODltih3niFGDQSIKNn0Q7iNF2Vp8FUzxiQHYHFawho4_2pIjoUJ8IcFsfbNr1BAZDtA8B3SmbK4hoSIeQpvAvcZNdduAi3v7tZfbx9Pi-esnXm-fX1cM633NGUy6wKDtWSao5VKh0XclaMlVyBoobUXFR6Yoy0wjQRmFZsEopXmtRSq6pLMQyuz975_5fE8bU9jbquTgM6KfYMs6ZbGjdsBm9O6P7mHxox2B7CIeWF1RKSYX4BU-wd-Y</recordid><startdate>20120925</startdate><enddate>20120925</enddate><creator>Duma, Mlungisi</creator><creator>Twala, Bhekisipho</creator><creator>Nelwamondo, Fulufhelo V.</creator><creator>Marwala, Tshilidzi</creator><general>Current Science Association</general><scope>7U1</scope><scope>7U2</scope><scope>C1K</scope></search><sort><creationdate>20120925</creationdate><title>Partial imputation to improve predictive modelling in insurance risk classification using a hybrid positive selection algorithm and correlation-based feature selection</title><author>Duma, Mlungisi ; Twala, Bhekisipho ; Nelwamondo, Fulufhelo V. ; Marwala, Tshilidzi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-j210t-3e45f1680c2a6ebc768781b521ab2d36236c601d93acdbe5416bb27c3582c0843</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Credit risk</topic><topic>Data imputation</topic><topic>Datasets</topic><topic>Information classification</topic><topic>Insurance risk</topic><topic>Machine learning</topic><topic>Missing data</topic><topic>Positive selection</topic><topic>Predictive modeling</topic><topic>RESEARCH COMMUNICATIONS</topic><topic>Sensors</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Duma, Mlungisi</creatorcontrib><creatorcontrib>Twala, Bhekisipho</creatorcontrib><creatorcontrib>Nelwamondo, Fulufhelo V.</creatorcontrib><creatorcontrib>Marwala, Tshilidzi</creatorcontrib><collection>Risk Abstracts</collection><collection>Safety Science and Risk</collection><collection>Environmental Sciences and Pollution Management</collection><jtitle>Current science (Bangalore)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Duma, Mlungisi</au><au>Twala, Bhekisipho</au><au>Nelwamondo, Fulufhelo V.</au><au>Marwala, Tshilidzi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Partial imputation to improve predictive modelling in insurance risk classification using a hybrid positive selection algorithm and correlation-based feature selection</atitle><jtitle>Current science (Bangalore)</jtitle><date>2012-09-25</date><risdate>2012</risdate><volume>103</volume><issue>6</issue><spage>697</spage><epage>705</epage><pages>697-705</pages><issn>0011-3891</issn><abstract>We propose a hybrid missing data imputation technique using positive selection and correlation-based feature selection for insurance data. The hybrid is used to help supervised learning methods improve their classification accuracy and resilience in the presence of increasing missing data. The positive selection algorithm searches for potential candidates for imputation and the correlation-based feature selection method searches for attributes have a significant effect on the target outcome. The imputation is performed only on those attributes that have an impact on the target outcome. The results show that the classification accuracy and resilience of supervised learning methods improve significantly when applied with the imputation strategy under these assumptions.</abstract><pub>Current Science Association</pub><tpages>9</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0011-3891 |
| ispartof | Current science (Bangalore), 2012-09, Vol.103 (6), p.697-705 |
| issn | 0011-3891 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_1221890791 |
| source | Jstor Complete Legacy; EZB Electronic Journals Library |
| subjects | Credit risk Data imputation Datasets Information classification Insurance risk Machine learning Missing data Positive selection Predictive modeling RESEARCH COMMUNICATIONS Sensors |
| title | Partial imputation to improve predictive modelling in insurance risk classification using a hybrid positive selection algorithm and correlation-based feature selection |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-20T18%3A24%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-jstor_proqu&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Partial%20imputation%20to%20improve%20predictive%20modelling%20in%20insurance%20risk%20classification%20using%20a%20hybrid%20positive%20selection%20algorithm%20and%20correlation-based%20feature%20selection&rft.jtitle=Current%20science%20(Bangalore)&rft.au=Duma,%20Mlungisi&rft.date=2012-09-25&rft.volume=103&rft.issue=6&rft.spage=697&rft.epage=705&rft.pages=697-705&rft.issn=0011-3891&rft_id=info:doi/&rft_dat=%3Cjstor_proqu%3E24088803%3C/jstor_proqu%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1221890791&rft_id=info:pmid/&rft_jstor_id=24088803&rfr_iscdi=true |