File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
  • Find via Find It@HKUL
Supplementary

Conference Paper: Hybrid Models Using Unsupervised Clustering For Prediction Of Customer Churn

TitleHybrid Models Using Unsupervised Clustering For Prediction Of Customer Churn
Authors
KeywordsChurn, Clustering
Data mining
Decision trees
Lift
Prediction
Issue Date2009
PublisherInternational association of Engineers, IAENG.
Citation
The Proceedings of the International MultiConference of Engineers and Computer Scientists (IMECS 2009), Hong Kong, 18 - 20 March , 2009, v. 1, p. 638-643 How to Cite?
AbstractIn this paper, we use two-stage hybrid models consisting of unsupervised clustering techniques and decision trees with boasting on two different data sets and evaluate the models in terms of top decile lift. We examine two different approaches for hybridization of the models for utilizing the results of clustering based on various attributes related to service usage and revenue contribution of customers. The results indicate that the use of clustering led to improved top decile lift for the hybrid models compared to the benchmark case when no clustering is used.
Persistent Identifierhttp://hdl.handle.net/10722/63308
ISSN

 

DC FieldValueLanguage
dc.contributor.authorBose, Ien_HK
dc.contributor.authorChen, Xen_HK
dc.date.accessioned2010-07-13T04:20:46Z-
dc.date.available2010-07-13T04:20:46Z-
dc.date.issued2009en_HK
dc.identifier.citationThe Proceedings of the International MultiConference of Engineers and Computer Scientists (IMECS 2009), Hong Kong, 18 - 20 March , 2009, v. 1, p. 638-643en_HK
dc.identifier.issn9789881701220-
dc.identifier.urihttp://hdl.handle.net/10722/63308-
dc.description.abstractIn this paper, we use two-stage hybrid models consisting of unsupervised clustering techniques and decision trees with boasting on two different data sets and evaluate the models in terms of top decile lift. We examine two different approaches for hybridization of the models for utilizing the results of clustering based on various attributes related to service usage and revenue contribution of customers. The results indicate that the use of clustering led to improved top decile lift for the hybrid models compared to the benchmark case when no clustering is used.-
dc.languageengen_HK
dc.publisherInternational association of Engineers, IAENG.en_HK
dc.relation.ispartofThe Proceedings of the International MultiConference of Engineers and Computer Scientist-
dc.subjectChurn, Clustering-
dc.subjectData mining-
dc.subjectDecision trees-
dc.subjectLift-
dc.subjectPrediction-
dc.titleHybrid Models Using Unsupervised Clustering For Prediction Of Customer Churnen_HK
dc.typeConference_Paperen_HK
dc.identifier.emailBose, I: bose@business.hku.hken_HK
dc.identifier.emailChen, X: chenxi@business.hku.hken_HK
dc.identifier.authorityBose, I=rp01041en_HK
dc.identifier.hkuros160496en_HK
dc.identifier.volume1-
dc.identifier.spage638-
dc.identifier.epage643-
dc.publisher.placeHong Kong, China-
dc.description.otherThe Proceedings of the International MultiConference of Engineers and Computer Scientists (IMECS 2009), Hong Kong, 18 - 20 March , 2009, v. 1, p. 638-643-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats