Discovery of Web Topic-Specific Association Rules
Web主题关联知识自学习算法

There are hidden and rich information for data mining in the topology of topic-specific websites. A new topic-specific association rules mining algorithm is proposed to further the research on this area- The key idea is to analyze the frequent hyperlinked relations between pages of different topics. In the topic-specific area, if pages of one topic are frequently hyperlinked by pages of another topic, we consider the two topics are relevant. Also, if pages of two different topics are frequently hyperlinked together by pages of the other topic, we consider the two topics are relevant. The initial experiments show that this algorithm performs quite well while guiding the topic-specific crawling agent and it can be applied to the further discovery and mining on the topic-specific website.