The Research Of Mining User Traversal Patterns Based On Web Log

With the development of Internet, WWW is applied in wider field. The action of the users can be known in Web Log. And this makes it possible to collect the information about the user traversal and analyse their action deeply. How to use these effective methods to find the useful information and make it invalid is a challenge. Then, the research on Web mining is brought forward and is becoming a hotspot.Data mining based on Web Log is a main aspect of Web mining. How to make the uesrs find the information they are interested in more quickly and expediently is the aim of every Web site. If the site's capability is improved, it will attract more users to visit it. And whether the site can provide the individuation service is an important factor to estimate it. Through data mining on Web log, we can find the user's traversal mode. It will help us to improve the site's structure and provide the better seivice to the users.This paper researches how to mine the user traversal patterns based on web log. And proposes a improved design of a system that can provide individuation serivice. The works are following:1) Study classical algorithms about sequence mode mining, analyse their theory and feature. Research the improvement and application of an algorithm in web log mining.2) Research a model of Web log mining. Predominate the whole process to carry out the mining. The process includes data mining preparation, finding the MFP and discovery of traversal mode. We use a simple example to carry out the data preparation, find out the MFP. Then present an improved algorithm to discover the frequent traversl path.3) Research on the application of Web log mining in Web site service. Proposed an improved system model of individuation service. The feature of this model is: (1) Using the data mining to find out what the user's interesting is and the traversal mode. This method is covert and can't be discovered by the user;(2) The way to commend the page to the user is different according to different user. Then the seivice is individuation and pertinence;(3) Discovery of the most close neighbour bases on the user clustering and this can make it more effective...