基于支持向量机的信用卡欺诈检测

Credit Card Fraud Detection Model Using Support Vector Machine

针对信用卡使用过程中存在的欺诈消费行为,运用支持向量机(Support Vector Machine,SVM)建立信用卡欺诈检查模型,以期取得较好的预测分类能力。本文从模型建立、模型评估、模型分析较为详细地介绍了模型的设计与实现过程,并将此模型分类结果与ID3+BP神经网络分类结果进行比较,验证了该模型的可行性和SVM分类的有效性。

A new detection method based on Support Vector Machine （SVM） was proposed in this paper. With the proposed scheme, the detection model for credit card was established with SVM. In this paper, the procedure tor designing and implementing the fraud detection mode was introduced in detail. The comparison between this mode and ID3＋BP （Back-Propagation） neural networks classification model was also performed. Test results show that model possesses preferable practicability and veracity.