Related Technology Research On Intelligent Web Page Recommendation System

With development and prevalence of Internet, more and more people gets the information and knowledge from Websites. As plenty of information are provided for us through Internet, it makes information search very difficult. How to help users to get the information what they need is a new problem which must be solved in further development of Internet. In this paper, based on the previously research, combining web content mining with web log mining, we present a new method to construct a intelligent web page recommendation system. This system may be used to recommend the user's interested web pages according his past access data.First, based on the analysis of access habit of user, we categorize user's access transaction into navigation path and interesting web pages set, and classify web pages into link pages and content pages. Second, general suffix tree algorithm is used to mine the key path from user's all navigation paths, and using key path to cluster all users into different interest groups. According interest group, we can make a primary recommendation pages set. To recognize the user's interests exactly, we construct the user's interested words sets from user's key path. At last, we defined the relativity of a web page to a user. It can be used to help us to select the web pages which the user needs truly.In this paper, we have given the system construction method and its implement algorithm in detail. In addition, we analyze the complexity of this algorithm. From the analysis, we can know this algorithm can be implemented easily.