Effective recommendation is indispensable to customized or personalized services. Collaborative filtering approach is a salient technique to support automated recommendations, which relies on the profiles of customers to make recommendations to a target customer based on the neighbors with similar preferences. However, traditional collaborative recommendation techniques only use static information of customers’ preferences and ignore the evolution of their purchasing behaviours which contain valuable information for making recommendations. Thus, this study proposes an approach to increase the effectiveness of personalized recommendations by mining the sequence patterns from the evolving preferences of a target customer over time. The experimental results have shown that the proposed technique has improved the recommendation precision in comparison with collaborative filtering method based on Top k recommendation.
Cheng, Tsang-Hsiang and Lee, Yen-Hsien, "IMPROVING RECOMMENDATION PERFORMANCE WITH USER INTEREST EVOLUTION PATTERNS" (2014). PACIS 2014 Proceedings. 298.