Credit card fraud data has high dimension and sparseness, the traditional detecting methods do not have good recognition rate. In order to improve the credit card fraud detection accuracy, this paper proposed a credit card fraud detection method based on support vector machine (SVM). Firstly, the credit card consumption data sam-pling is trained by support vector machine to build a detection system, then the detecting system credit card consump-tion behavior is tested to determine whether the fraud trading behavior is fraud. The model is tested by a commercial bank credit card consumption data, and experimental results show that the credit card fraud detection accuracy reached 95% using support vector machine, with testing time only 0. 565 seconds. The results prove that the pro-posed detection is an effective credit card detection method%研究信用卡安全优化设计问题,信用卡欺诈数据具有高维数和稀疏性,由于欺诈样本数据的冗余特征,导致传统检测方法不能很好的识别信用卡欺诈行为,导致检测准确率低.为了提高信用卡欺诈检测准确率,提出一种支持向量机的信用卡欺诈检测方法.首先用采样来的信用卡消费数据训练好一个支持向量机检测系统,然后用支持向量机检测系统对一父信用卡消费行为进行检测,判断是否为欺诈交易行为.对某商业银行的信用卡消费情况进行测试实验,实验结果显示,采用支持向量机的信用卡欺诈检测精度达到95％以上,且检测时间只有0.565秒,说明提出的检测是一种有效的信用卡检测方法.
展开▼
