A New Multi-Criteria Quadratic-Programming Linear Classification Model for VIP E-Mail Analysis

Download PDF
Open Webpage

Peng Zhang, Juliang Zhang, Yong Shi

Published: 2007, Last Modified: 21 Jan 2026International Conference on Computational Science (2) 2007EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: In the recent years, classification models based on mathematical programming have been widely used in business intelligence. In this paper, a new Multi-Criteria Quadratic-Programming Linear Classification (MQLC) model is proposed and tested with VIP E-Mail dataset. This experimental study uses a variance of K-fold Cross-Validation to demonstrate the accuracy and stability of the model. Finally, we compare our model with the decision tree by using commercial software C5.0. The result indicates that the proposed MQLC model performs better than decision tree on small samples.