The Loss of Consumers´ Analysis Model Based on Strong Attributes Restricted Bayesian Classifier

With the increasingly competition in E-commerce, the analysis of the loss and value of consumers has become a key problem to the enterprise organization. A strong attributes restricted Bayesian classifier is described in the paper, the SANBC which is a restricted Bayesian classifier based on strong attributes extends the structure of the naïve Bayesian classifier through the adding of highlighting lines between strong and weak attributes so that the naïve Bayesian classifier can be weakened. The classifier could build a mathematics model to analyze the loss possibility based on the attributes of consumers, service, and the consume record of client. It is an effective research in client relation management for E-commerce.