Extraction and Credibility Evaluation of Web- based Competitive Intelligence

摘要

Web has been one of major information sources for enterprises to acquire competitive intelligence. However, traditional approaches focus on collecting Web pages and fail to generate practical competitive intelligence from Web pages. Another problem in the research on Web-based competitive intelligence is that Web pages may contain a lot of incredible information which will have big influence on the effectiveness of competitive intelligence. Aiming at solving these problems, we propose a framework in this paper for the extraction and credibility evaluation of Web competitive intelligence. We present an entity-based approach to extracting Web competitive intelligence, and a social-network-based method to evaluate the credibility of acquired competitive intelligence. The entity-based extracting approach is based on an ontology of Web competitive intelligence, which represents competitive intelligence as a set of competitor intelligence and competition environment intelligence. Some critical issues about the entity-based approach and the social-network-based method are analyzed in detail. The results show that our system is useful to improve the effectiveness of the extraction and credibility evaluation of Web competitive intelligence.