关键词: 信息内容识别, 刷单现象, SVM算法, K折交叉验证

Abstract: Although the click farming of e-commerce platform improve the store revenue to some extent, click farming on the one hand raised the cost of promotion of e-commerce platform causing a serious problem of reputation security. On the other hand, false information has led consumers to be easily misled resulting in property loss. In the view of the phenomenon of e-commerce platform click farming, this paper introduced an intelligent method(SVM-NB algorithm) of detecting the click farming of e-commerce platform for the users and proposed a means of constructing the characteristic values of click farming. Firstly, it will collect the relevant data of commodity and create an eigenvalue database. Then it will establish a classifier based on Support Vector Machine (SVM) algorithm with supervised learning so as to solve the result of the judgment of click farming. Finally, the system will calculate the probability of the click farming of goods adopted Naive Bayes(NB) which can be provided shopping reference for consumer to help them choose more valuable goods. In this paper, the validity and accuracy of SVM-NB algorithm are verified by k-fold cross validation algorithm. The accuracy of the calculation results is as high as 95.9596%。

Key words: Information content identification, phenomenon of click farming, Support Vector Machine algorithm, k-fold cross validation algorithm