Multiple criteria programming models for VIP E-Mail behavior analysis

Publisher:
IOS Press
Publication Type:
Journal Article
Citation:
Web Intelligence and Agent Systems-An international journal, 2010, 8 (1), pp. 69 - 78
Issue Date:
2010-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2011000667OK.pdf208.62 kB
Adobe PDF
Excessive lose of customer account is becoming a major headache for VIP E-Mail hosting companies. Analysis of what kind of customer is more prone to lose and finding the appropriate measures to sustain those customers has become urgent needs. Recently, classification models based on mathematical programming have been widely used in business intelligence. The purpose of this paper is to propose several multiple criteria programming methods for classification and apply these methods to VIP E-Mail behavior classification. We first introduce a model for a generalized multiple criteria programming based classification method, specifically four particular forms, and then we use a cross-validation method to test the stability and accuracy of multiple criteria programming methods on VIP E-Mail accounts. Finally, we compare our models with Support Vector Machine (SVM). The results show that the classification models based on mathematical programming are satisfactorily accurate and stable on a VIP E-Mail dataset. Therefore, it can be concluded that applying the proposed method on VIP E-Mail behavior analysis can provide stable and credible results.
Please use this identifier to cite or link to this item: