Research on Customer Value Classification Based on Improved Support Vector Machine

摘要

Aimed at the shortages of the existing data-mining model for classification of customer, this paper proposed a new customer classification model based on rough sets and support vector machine. First, the theory of rough set was applied to pick up and reduce the index attributes. Then, the training samples were sent to the support vector machine to train and learn. After that, the sorts of the customers in test samples were determined. The test results indicate that the new customer classification model based on rough sets and support vector machine shows the higher forecast precision than the traditional customer classification models and it is more efficient and practical.