User Preference Collaborative Filtering Recommendation Algorithm based on Data Mining

With the rapid development of information technology, the Internet has developed into the most important e-commerce platform. This article integrates user preference mining technology into collaborative filtering recommendations and proposes an e-commerce collaborative filtering recommendation algorithm based on user preference mining. This algorithm Aiming at the traditional collaborative filtering recommendation algorithm that only uses the user's explicit preference information when calculating user similarity, and ignores the user's implicit preference knowledge, it is proposed to use user preference mining technology to perform user explicit preference knowledge and implicit preference information. Mining preference knowledge, using the excavated user preference knowledge to calculate user similarity, and realizing the nearest neighbor community formation mechanism based on user preference knowledge. On this basis, intelligent recommendation of user needs is realized. Experiments show that the algorithm has achieved expectations Effective, comprehensive use of user preference knowledge for collaborative filtering recommendation is the key to improving the accuracy and quality of recommendation results.