Study Of The Access Pattern On Web Page

The paper introduces the definition, basic process, categories and general technology of Web mining. Moreover, Difficulty of Web Mining has been shown by means of comparing three kinds of methods. The paper discusses the problem in the process of building personal web site and confirms the object with a conclusion of personal web site research direction.The paper researches the user's behavior; makes use of the balanceable interest that reflects by high frequency access paths. Makings use of the technique of Web Mining, automatically discovers the users access pattern from the Web Log file - high frequency access paths; According to the page that current users request, find out the high frequency access paths which are match most in current time, forecast the page that users next step will browse, forwardly adjust Web content and increment users' cache in order to shorten responding time, make users get mostly interested content. The main idea is: (1) Web Log mining to get high frequency access path, it reflects the sequence that fits most users. Several of them are very likely interested for the current users can represent to further browse behavior. (2) if mostly users' access sequences are just for getting the page dn in high frequency access path (d1,d2,.. .,dn), then the dn be very likely really interested, can directly send dn to user.The paper constructs a real system of PWS based on the interesting. The whole system is made up of four parts: system management, pretreatment, access pattern mining and personal recommendation. The system real time monitors users' behavior. Then the systems, according to the Log, mine the high frequent path and dynamically recommend page that have high interest. At the same time the system sends the page to the catch to economize users' time and money. Besides, I have constructed real application to complete and supply the technology, at last, test it in theory and practice.Use access pattern mining out from the Web Log to recommend the pages that the users need, then provide the characteristic service for the users, improve the system's performance. This system functions on the service part and transparent to users. Through managing part can control the system; make the system have the higher adaptability.