基于SVM的Web日志挖掘及潜在客户发现

潜在的客户资源是商家未来的利润来源，发现了潜在的客户就可以制定相应的商业决策，并进行有针对性的客户关系管理。使用SVM方法对Web日志文件进行挖掘，以发现站点访问者中潜在客户的共同行为模式，并将其分为不同级别的目标客户群。同时，通过试验4种不同比例的训练样本，研究了非对称数据对分类结果的影响，以期获得较优的模型。

Application of SVM in Mining Potential Customers from Web Log

Potential customers are the future sources of profits.The enterpriser can make business decisions accordingly,and can manage customer relationship specifically as soon as finding those people.In this article,support vector machine(svm)is used in web log mining,in order to search common patterns of behavior from potential customers who visit the web sites.And those potential customers are divided into different classes.Then tests on four train sets with different proportion present the influence of asymmetric data sets on classification results,and give a better model finally.