A Collaborative Filtering Algorithm Fusing Category Preference and Spatiotemporal Information

In order to overcome the disadvantages of the traditional collaborative filtering algorithm and reduce the user interest caused by the rarity of user interest matrix, this paper proposes a novel collaborative filtering algorithm based on user experience, which integrates category preference and spatiotemporal information. The algorithm combines the similarity of category preference with the traditional similarity, and introduces the time and location factors into the collaborative filtering algorithm. In addition, the algorithm also analyzes and calculates the time similarity and location similarity, and uses the comprehensive weight to obtain the user similarity. The research results show that the algorithm breaks through the limitations of the traditional collaborative filtering algorithm and fundamentally improves the accuracy of recommendation.