摘要
为了提高电子商务移动支付风险预测精度,针对当前电子商务移动支付风险预测建模过程存在的一些局限性,设计了粒子群算法和支持向量机的电子商务移动支付风险预测模型(PSOSVM)。首先收集大量的电子商务移动支付风险预测历史样本数据,并对其进行预处理,得到训练样本和测试样本集,然后采用粒子群算法和支持向量机联合对训练样本集进行学习,建立可以描述电子商务移动支付风险变化特点的模型,最后采用VC++编程实现电子商务移动支付风险预测仿真测试。结果表明,PSO-SVM克服了当前电子商务移动支付风险预测模型存在的弊端,电子商务移动支付风险预测精度超过90%,而且预测结果稳定可靠,相对其他电子商务移动支付风险预测模型,具有显著优越性。
In order to improve the accuracy of e-commerce mobile payment risk prediction,a particle swarm optimization(PSO-SVM)and a support vector machine(SVM)model for e-commerce mobile payment risk prediction are designed to overcome the limitations of the current e-commerce mobile payment risk prediction modeling process.Firstly,we collect a large number of historical sample data of e-commerce mobile payment risk prediction and preprocess them to get training samples and test samples.Then we use particle swarm optimization and support vector machine to learn the training samples,and establish a model that can describe the characteristics of e-commerce mobile payment risk change.Finally,we use VC++programming to realize e-commerce mobile.Payment risk prediction simulation test results show that PSO-SVM overcomes the drawbacks of current e-commerce mobile payment risk prediction model.The accuracy of e-commerce mobile payment risk prediction exceeds 90%,and the prediction results are stable and reliable.Compared with other e-commerce mobile payment risk prediction models,PSO-SVM has significant advantages.
作者
郝柯羡
HAO Ke-xian(School of Humanities and Economic Management,Xi’an Traffic Engineering Institute,Xi’an 710300,China)
出处
《电子设计工程》
2020年第15期79-82,87,共5页
Electronic Design Engineering
关键词
粒子群优化算法
电子商务
移动支付
风险预测
仿真测试
particle swarm optimization algorithm
e-commerce
mobile payment
risk prediction
simulation test