Research And Implementation On Internet-based Bank Competitive Intelligence Collecting System

Recently, the process of enterprises informatization in enterprises has been enhanced by the rapid development of Internet. Meantime, enterprise competitive intelligence system at the core of enterprise informatization process is also faced with the challenge of the rapid development of Internet. In this context, the study that how intelligence be collected, analyzed and presented automatically, accurately and effectively is of important research value and significance toward the new competitive environment.Distinguishing the primary content blocks in web page is the base of computing topic relatedness and the accuracy rate of intelligence collecting is affected by this technique to a large extent. Based on the existing techniques, this paper introduced a dynamic vote strategy and content analysis technique to improve the accuracy rate and recall rate for distinguishing the primary content blocks in web page. The experimental results showed that the solutions proposed in this paper were better than the present approaches in performance.The key to automatic intelligence collecting techniques is evaluating the information to be analyzed automatically to judge whether it is the needed intelligence or not. So this paper proposed an evaluation model for degree of topic relatedness based on dynamic matching after carrying out a substantial amount of research, which is able to improve the speed of evaluating topic relatedness for web page without reducing the accuracy rate in user's view.This paper analyzed bank competitive intelligence and enterprise competitive intelligence system. On the basis, an overall framework of enterprise competitive intelligence system was proposed in the paper. In the end, a bank competitive intelligence collecting system was designed and developed which is able to collect bank competitive intelligence automatically.