This thesis studies a user interest model which can correspond with every user by collecting some personal information of users and optimize the retrieval results by search the original query expansion automatically. After putting the inquiry keyword into the search engine, the system will filter and sort the retrieval results according with the relevant of the user's interest model and finally present it to users. The experiments show that using this method can improve the precision ratio and recall ratio and realize the personalized service of the information index for users.
展开▼