고객반응 이력과 상품정보를 반영한 군집화 기반의 상품 추천 모델

The proposed model seeks to solve the problem of the existing recommendation algorithms in which products to be recommended are processed within a limited group and products with high performance are repeatedly recommended due to the bias toward products with high user preference. This study utilized K-means, which is the most actively used clustering learning algorithm for recommending products to customers in online shopping malls. After creating two clustering categories and applying collaborative filtering, the customer's response history to the product is calculated according to the weight, and then the final recommended product is randomly selected. Through this model, more diverse products can be recommended to customers. This study is, a wider range of products will be recommended to customers in online shopping malls to broaden the range of product choices for customers and to help companies show an even distribution of product sales. A case study was conducted using the data of ‘A’, an online shopping mall currently in operation. The algorithm of the proposed model describes the process of selecting 10 recommended products when a customer checks the detailed information of a specific product.