The research and implementation of a correlative degree mining algorithm based on IIS logs

In order to find out the user patterns that hide in Web logs, log mining technology is one of the best ways. Log mining is the usage of data mining in the field of Web server´ logs. Although there are a set of softwares which can be used to analysis Web logs, the algorithm raised in this article pay special attention to discover the relationship among all the pages of the Web site. In this algorithm, size-link radio and static inner-link degree was creative used. According to the result of experiment, this algorithm can exactly find out the correlative ones among massive pages.