DSpace Repository

Integrating boosting and genetic algorithm with supervised classification for socio-demographic based customer targeting

Show simple item record

dc.contributor.advisor Vatcharaporn Esichaikul (Chairperson) en_US
dc.contributor.author Tutiyaporn Nitichai en_US
dc.contributor.other Donyaprueth Krairit (Member) en_US
dc.contributor.other Guha, Sumanta (Member) en_US
dc.date.accessioned 2015-01-12T10:43:42Z
dc.date.available 2015-01-12T10:43:42Z
dc.date.issued 2009-07 en_US
dc.identifier.other AIT RSPR no.IM-09-10 en_US
dc.identifier.uri http://www.cs.ait.ac.th/xmlui/handle/123456789/471
dc.description Submitted in partial fulfillment of the requirements for the degree of Master of Science. en_US
dc.description.abstract Naïve bayes is a supervised classification technique which aims for classifying or predicting group to unknown data. It assumes class-feature independence and no conditional dependency between classifying features. Generally, this technique delivers result with 60%-90% accuracy. Boosting algorithm is used as an ensemble technique for improve an accuracy of classification algorithms. Genetic algorithm is a promising technique for solution optimization and complex problem. In data mining aspect, genetic algorithm is used in feature subset selection which intends to find a set of predicting features that influence the most to the class label. This paper presents a combination of boosting and genetic algorithm with Naïve bayes in order to increase the prediction accuracy. The experiment is operated to prove that the integration of genetic algorithm, Adaboost with Naïve bayes can improve Naïve bayes’s prediction accuracy. This research runs the experiment by applying the integrated technique over a socio-demographic data set provided by an insurance company. Genetic algorithm is applied for feature selection and Adaboost is utilized for ensemble of Naïve bayes as its based classifier. The company aims to analyze this data set for customer targeting. The result shows that the integrated techniques slightly improve the prediction accuracy. This paper also discusses the reasons causing small improvement. en_US
dc.description.sponsorship Royal Thai Government (RTG) en_US
dc.language.iso eng en_US
dc.publisher Asian Institute of Technology en_US
dc.subject Customer services en_US
dc.subject Data mining en_US
dc.subject.lcsh Others en_US
dc.title Integrating boosting and genetic algorithm with supervised classification for socio-demographic based customer targeting en_US
dc.type Research Report en_US
dc.rights.holder Copyright (C) 2009 by Asian Institute of Technology. en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse

My Account