Research On Meta-Search Engine Results Sort Algorithm Based On User Interest Model

With the rapid development of Internet, the amount of resources on Internetcontinued explosive growth. Internet is highly open and heterogeneous distributedinformation space which without unified management. Information scattered distributed invarious sites around the world and updating quickly every day. The search results oftraditional search engines contain a lot of information which users are not interested in,even repeat or ineffective messages. Each user has different demand for informationretrieval. Traditional search engines often do not consider the difference between differentusers when providing search services. It is a time-consuming and labor-intensive work foruser find the information really needed. The Meta-Search Engine which Established basedon traditional search engines has the same drawback, although the meta-search enginecompared to the traditional search engine to a certain extent enhance thecomprehensiveness in querying information. A kind of personalized search service basedon users’ different interests is urgent need. Considering of the results processing featuresof the meta-search engine has much room for optimization. It can make a goodcombination with the personalized service. So this article does the work basedon previous research on personalization and meta-search engine, making a combinationand provides a kind of personalized meta-search engine.On the basis of previous research, by introducing the ontological thinking intopersonalization we provide a personalized meta-search engine to compensate for thepersonal limitations of traditional meta-search engine. Firstly we introduce the researchstatus of user interest modeling and based on the previous research and provide a kind ofuser modeling method with ontological thinking and describe how to construct the usermodel. Then we put forward a kind of forgotten the reconstruction mechanism basedon the forgotten model of psychology memory for user interest model updating. The user’sinterest was divided into long, medium and short-term different interest categories tovarying degrees of amnesia. And the interest in the class of semantic association isconsidered during the update processing. Finally ontology-based user interest model is applied to the meta-search engine by combine the user interest model with the QueriedResult Ranking Function in order to achieve personalization features.