In this paper we present a system framework for the extraction of intelligence about competitor from the Web. With the surprising increasing of the data volume in the Web, how to get useful intelligence about competitor has been an interesting issue. Previous study shows that most people prefer to look up information by competitor. We first analyze the requirements on the extraction of competitor intelligence from the Web and define three types of intelligence for competitor. And then a system framework to extract competitor intelligence from the Web is described. We discuss the three key issues of the system in detail, which are the profile intelligence extraction, the events intelligence extraction, and the relations intelligence extraction. Some new techniques to deal with those issues are introduced in the paper.
展开▼