The Research Of Personalized Recommedation Service Based On Web Usage Mining

With the rapid development of the Internet applications, there is a sharply increased demand on information services via the Internet. While the huge amount of information distributed on the Internet makes it harder for the individual user acquire what he or she needs. Such a phenomenon could be called information overload. At the same time, the distribution of information on the Internet makes the individual user find useful one more difficult, that is so-called information astray. Being insufficient in positive characteristic and lack for taking users' interest into consideration, most search engines presently have not yet solve problems including information overload and information astray. Until recently, the web personalization service has been proposed so as to settle down those issues.The web personalization service based on Web usage mining is the key technique in both researches and applications. The web usage mining could be applied in the personalization service, the commercial intelligence, and improving the structure of the websites. In the paper, the key technologies in personalization recommendation service based on web usage mining were studied in detail.The main content in the paper is as follows:1. The web usage mining was studied completely, including data collection, data preparation, pattern discovery, pattern analysis, and applications.2. A scalable interest model (SIM) was proposed. Firstly, some methods to present and the measure user's browsing interests and CI (cluster's interest characteristic) in the present personalization service systems were introduced and their deficiency was analyzed. Then, a scalable interest model (SIM) was proposed. The SIM makes use of user session's 2-segment characteristic to present SI (session's interest characteristic). The SIM is a scalable model. It could be adaptive to the characteristic which was extracted from 2-segment so as to adjust the accuracy of the presentation of SI and CI. The experimental result showed that the SIM improved effectively the accuracy of both SI and CI.3. A personalization recommendation algorithm based on session cluster (SCRec) was proposed. Firstly the deficiency of the personalization recommendation based on cluster mining technology was analyzed in detail. Then a personalization recommendation algorithm based onsession cluster (SCRec) was proposed. During the stage of data preparation and clustering, SIM could be used to present SI and CI. During the recommendation stage, the weight of 2- segment in the session and the similarity of the cluster could be used for recommendation. The experimental result showed that SCRec could enhance the service quality of the recommendation system effectively.