Application of E-Commerce Recommendation Algorithm in Consumer Preference Prediction

Through user characteristic information, user interaction behavior, commodity characteristic information, recommendation engine, and related technologies in data mining, this paper makes a more in-depth study and analyzes the problems of "big data volume," "cold start," and "data sparsity" in the recommender system in modern business websites. In response to these problems, this paper transforms the problem of large data volume into the problem of large user groups. Then, after using the k-means clustering algorithm to divide the large user group into homogeneous user groups to alleviate the problem, a combination of collaborative filtering algorithm and content-based recommendation algorithm in the homogeneous user group is proposed to alleviate this problem. The experimental precision and recall are both around 0.4, and when W=0.8, the F value is the largest.