User Interest Model Update Mechanism in Personalized Information Retrieval

摘要

In order to better realize the personalized information retrieval service, the key is dynamically updating user interest model. This paper puts forward a user interest model based on ontology, First by analyzing the user browsing behavior in the Web environment, we improve the calculation method of interest right value. And then, according to the characteristics of the web document to improve the TF-IDF algorithm, in this foundation we propose the user interest model updating and forgetting mechanism. The experiments show that the method can capture the change of user interests, further improve the accuracy of personalized information retrieval and user satisfaction.