The Research Of Ontology And Its Usage In Personalized Information Retrieval

With the development of the Internet, information was mounted up exponentially everyday. It becomes more and more difficult for the user to find his favorite information in the numerous and jumbled network. Nowadays, the keywords-based retrieval technique can meet the user's needs far from. The received information is still in a mess by keywords retrieval. To find the information he wants, the user must spend a lot of time and energy. This makes the user meeting obstacles here and there during the process of taking full advantage of network resources.Personalization and specialization is becoming the most important component of the web search. To satisfy the user's needs and help the user to find information he needs from numerous and jumbled network information. Making the retrieval service conforming to other services and providing the user systematically, comprehensive and accurate information services is becoming the main tendency of search engines. Currently, the major search sites, such as Yahoo, Google, Baidu and china searches, provide specialization query services. They classify the net resource according to subject in advance. While the user queries, he can choose the corresponding subject firstly and then input the keyword to search for. In addition, the china searches web site also provides intelligent navigation after keywords search. (For example, when we input keyword "computer", we will have the following navigation information: software download/ company/education/school academic teaching system). Search results will be reclassified by the chosen navigation information. This can reduce the redundancy of information and enhance the retrieval efficiency to someextent. However, it still has a great gap to the personalized retrieval we hope for.The aim of personalized service is to provide different services to satisfy different needs of different users under different background at different time. The linchpin of the personalized service is to construct user profiles.Combined with the research of ontology and personalized service, the concept of user ontology in personalized retrieval was proposed in this paper. The construction and maintenance of user profile based on behavior was introduced. And the extraction rules of semantic relations were discussed in detail. The method of construct the user ontology was given based on the discussions. Finally, the model of the user ontology-based personalized information retrieval systerm was given . A kind of multi-database knowledge acquriation systerm model was discussed for the users need in all kinds of forms the knowledge appeared.