C2C E-Commerce Recommender System Based on Three-Dimensional Collaborative Filtering

摘要: To solve the special recommendation problem in C2C e-commerce websites, a three-dimensional collaborative filtering method which can recommend seller and product combinations is proposed by extending traditional two-dimensional method. And recommender system based on designed. The framework of key calculations process are discussed. firstly calculates similarities using features, fills rating set sales relations to sparsity data. Then it buyer historical ratings, decides neighbors predicts unknown ratings. Finally recommends with highest prediction ratings target buyer. A true data experiment proves good performance system.