The rapid growth of World Wide Web in recent years has made it important to carry out resource discovery Topic specific web crawler collects relevant web pages of interested topics from the lnternet, there are many relevant researches focusing on topic-specific crawling. However few works detail the topic-specific crawling with the user interests. In this paper, we present a new user interests model to optimize the performance of the topic-specific crawler. The crawler can learn from the previous experience to improve the proportion of the number of relevant pages and the number of the whole pages by using the user information, which is collected by data mining approach.
展开▼
