Personalized Recommendation System for Offline Shopping

This paper studies and establishes a system for the problem of lack of personalized commodity recommendation and low pertinence in shopping offline. Compared with online recommendation, offline system has inherent disadvantages on data. This paper overcomes its difficulties and does a further analysis and research on shopping information and commodity image of offline stores, and the algorithm model for offline personalized intelligent recommendation system was established, then the system was constructed to demonstrate its practicability and feasibility. Finally, we described the future of offline intelligent recommendation system and the difficulties to be solved, also we provide a promising outlook.