Applications Of Web Mining In Personalized Search Engine

With the development and popularity of Internet, Web has become an important way of getting information. Because of the rapid expansion of online information, SE plays a very important role in searching network information. Web mining provides technical support for personalized information search. Personalized service oriented towards user profiles has become a hot issue of research in Web technologies. Researching on applications of web mining in personalized search engine, the main tasks for this dissertation are as following:Firstly, this dissertation analyzes the current status and problems of SE, points out the significance of studying personalized information retrieval. It analysizes the working principle, the organization structure and functions of SE. Detailed studies are also made on the Space Vector Model (VSM).Secondly, this dissertation introduces the definition and tasks of Web mining, studies and analyzes the relation between Web mining and traditional information retrieval and the applications of Web mining in SE. It introduces in detail the methods of accessing users interests based on users browsing behavior on Web. This dissertation provides a method of improving VSM. It studies the key technologies which aim to use VSM to establish initial user interest models, and then automatically understand and learn the interest of users based on the mining behaviors of users; and then renew the models with filtered personalized information. In addition, due to the existence of the large number of synonyms and in order to improve retrieval coverage, it is recommended to establish synoymsdictonary to expand the search and optimize the query.Finally, based on the above studies, this dissertation designs a personalized information retrieval PSE based on Web mining, and describes in detail the functions and realization of related modules. It carries out study from a theatetical prospective about the fact that the personalized SE has the characteristics such as flexbile searching methods and personalized retrieval results etc. At the end of this dissertation, the author summarizes the work that has been done and explores the development potential of personalized information retrieval.