Literature Recommendation Technology Based on User Characteristics and Semantic Computing

Obtaining the firsthand literature resources related to research is one of the main tasks of intelligence personnel. Different from the open users to whom content recommendation service is oriented, intelligence personnel have a clear professional background and specific research needs. This paper proposes a personalized recommendation algorithm for scientific literature that integrates user characteristics and semantic computing. On the one hand, using the rich background information and scientific research experience of intelligence personnel were used to generate recommendations for static information modeling; on the other hand, user behavior was used to explore the potential semantic relationship between the literatures. Finally, the two methods are combined. Through the effective integration of user static characteristics and user behavior information, the dependence on the user interaction behavior was weakened, and the cold start problem was effectively alleviated. The experimental results showed that the proposed method was significantly improved compared with the traditional method, which proved the feasibility of the method.