Research On Intelligent Recommendation System Based On Web Log Mining

With the developing of network technology, the Internet plays a more and more important role in many aspects. Then the personalized information service is much more attractive because of the plenty of resources on the Internet. Intelligent Recommendation System (IRS) is born in this environment.IRS uses data mining and AI technology, provides intelligent and personalized service. It could recommend pages or products to the customers, and could lead customers to focus on the information that they want.The core of IRS is the recommendation method. We find that Markov model is convenient and easy for prediction. So we find the way to use Markov Model in the IRS. When we study the method of Markov in prediction, we find it's time and space cost is very large. In order to solve the problem, we use a novel method including Markov Model, clustering, page weight and individual bias. The new method use clustering to reduce the cost of computing Markov matrix, and compute the page weight and individual bias to improve the hit rate .At the end of the thesis, we build a framework of IRS using multi-agents, and use the Markov method to introduce the functions of the main agents.