Selecting Salient Features and Samples Simultaneously to Enhance Cross-Selling Model Performance
Dehong Qiu, Ye Wang, Qifeng Zhang · IGI Global eBooks · 2010
The task of the 2007 PAKDD competition was to help a finance company to build a cross-selling model to score the propensity of a credit card customer to take up a home loan. The present work tries to increase the prediction accuracy and enhance the model comprehensibility through efficiently selecting features and samples simultaneously. A new framework that coordinates feature selection and sample selection together is built. The criteria of optimal feature selection and the method of sample selection are designed. Experiments show that the new algorithm not only raises the value of the area under ROC curves, but also reveals more valuable business insights.