Based On The User Model Of Personalized Search

With the continued development and popularity of the Internet technology, the technology of the information access has become an indispensable part of the people’s daily life. In this context, every user is both the provider and the customer of the information, which results in all kinds of information coming into being on the internet and a sharp increment about the volume of the information. As to the access of the information, the traditional methods of the search engine can’t meet the user’s individual requirements without the distinction of the searching results, so the technology of the search engine based on the user personalization becomes the inevitable trend in a certain period. Therefore, the paper uses the information of the users’ interests to build the interest model and apply it in the process of the information retrieval. Besides, carries out the related research about the personalization search based on the user interest model. The main contents of the paper are as follows:Firstly, the paper introduces the working principles of the personalization search engines and studies the relative theories about the information retrieval, especially the key technologies about the search engine based on the user personalization, including the vector space model (VSM) and its important application in the document matching, such as the calculation of the document similarity.Secondly, does in-depth research about the user interest model, discusses the key technologies about the procedure how to access the user interest information, how to express and model these information. Based on this, we design a new user model which combines the explicit feedback and the implicit feedback. At the same time, improves the traditional implicit modeling method, such as the consideration of the part of speech of the vocabulary about the user interest when assign the weight. Then, give the methods about the initialization, establishment and update of user interest model with the combination of the explicit and implicit feedback.Finally, realizes one experimental system based on the Lucene architecture and analyzes the concrete implementation mechanisms from the personalization system’s structure, process, sorting, scoring and some other aspects. In order to verify the effectiveness, we design some related experiments to test the performance of algorithm proposed in this paper and analyze the corresponding results.In summary, the paper proposes a new information search model based on the user interests, and expects provide important reference for the research and application of the information search technology.