SVM在移动通信客户流失预测中的应用研究

使用支持向量机(SVM,Support Vector Machine)数据挖掘方法对移动通信行业客户流失倾向进行预测,对支持向量机同决策树算法预测的结果进行对比,结果表明支持向量机对本文所选取的属性数据具有更强的分类能力,而且在不同训练数据规模情况下预测模型有较好的稳定性。实验证实,运用本研究模型选取全体客户的22.31%,可以预测出50.07%流失的客户,表明本研究中提出的预测模型具有实际应用价值。

Research on the Application Prediction of Cell Phone churn based on support Vector Machine

Data mining based on support vector machine(SVM) is applied to prediction of churn of cellular phone customer. The model is also compared with the models based on decision tree. The experimental results on the real customer data demonstrate that the model based on SVM achieves good performance and has less prediction errors than those of decision tree models and has more stability. The empirical results showed that the prediction mode was effective in predicting at risk churn of cellular phone customer. The proposed model could identify 50.07% churners by selecting 22.31% of the customer.