Based On Data Mining, Web Mining System

With the explosive growth of information on the World Wide Web, it is necessary to provide users with tools for efficient knowledge discovery on the Web. The target of the search engine is not knowledge discovery but information retrieval. With the application of data mining to the Web, Web mining has become one of the most important and flourishing fields of data mining.This paper emphasizes on the design of Web mining system. A current prototype of Web mining system Web_Ms is presented. The purpose is to provide a practical model for Web mining and make it more effective to discovery knowledge on the Web.To deal with the semi-structured data on the. Web, this paper transforms the semi-structured data to well-structured data by extraction based on XML and builds a multilevel Web database. Meanwhile, it preprocesses Web logs. It solves the problem of data standardization for Web mining system. On how to implement the function of mining, this paper takes the strategy that integrates different kinds of methods for mining into a warehouse and inserts it into the system as a module. The module has its own interface and method to be used. It cooperates with other modules to complete the task of Web mining.The prototype of Web mining system Web_Ms not only provides a valued model for developing tools for Web mining, but also promotes the research of Web mining in turn.