FCA - Based Web User Profile Mining for Topics of Interest

To tackle the problems of Web information overload and mismatch, researchers have proposed many approaches to acquire Web user profile for topics of interest. But the traditional techniques do not provide satisfactory structures for mining Web user profile. In this paper, we present a new approach automatically mining Web user profile based on FCA (formal concept analysis) methods from positive documents. We use the formal concepts with their weights as patterns to represent topics of interest. Based on the patterns discovered, the process of assessing documents relevance to the topics is also proposed. The experiments show that our approach is effective.