A Weibo Recommendation Algorithm Integrating User Interests

Abstract

A Weibo recommendation algorithm integrating user interests is studied in this manuscript. Among the algorithms based on content similarity recommendation, the LDA algorithm is one of the most used and classic algorithms. Therefore, this paper uses the LDA algorithm to mine the user interest distribution and uses the cosine similarity algorithm to calculate the similarity between the general microblog to be recommended and the user's interest. The proposed algorithm is designed by considering the similarity of user interest based on the semantic level and the similarity of the social relationship based on the follow relationship are then considered, a better recommendation effect can be achieved. To validate the model, the proposed algorithm is simulated on the data collected from the Weibo. The accuracy is higher than before.