Expression of user personalized search behavior based on keyword query series and Bayesian Network

Nowadays the existing search engines are always lack of the consideration of personalization. They display the same search results for different users despite their differences in interesting and purpose. In order to solve this problem, this paper introduces a new method of using keyword query series to express the personalized search behavior of a user. Based on the keyword query series we construct a keyword query graph for every user. Also using Bayesian network, we construct the prior probability of keyword selection and the migration probability between keywords for every user. Then we can calculate the similarity between every two users in order to do the recommendation based on neighbors. In this way, we construct a dynamic and personalized search behavior profile for each user, which can be used to build a personalized search engine.