Collaborative Filtering Recommendation Algorithm Via Sparse Data Optimization

With the development of e-commerce recommendation system,as one of the most widely used personalized recommendation algorithms,the traditional collaborative filtering recommendation algorithm faces a series of problems such as data sparsity and cold start.In this paper,some improved methods are proposed to solve these problems.The main contributions of this paper is as follows:(1)A sparse data preprocessing method based on item attribute has been proposed.The improved sparse data preprocessing method combines the user history scoring and item attribute characteristic factors.First,considering the item characteristics information,the ratings of the unrated items are predicted through the similarities between each item.It can lead to saturated matrix and overcome the drawback of the sparsity matrix.Next,the hybrid filling method is utilized to process the unrated items in the sparse data sets.It can avoid the problem of full no consistency of different items for traditional mean-filling method and the multiple mode and no mode for the mode-filling approaches.The simulation results demonstrate that the proposed algorithm can improve the recommended quality dramatically,compared with existing methods.(2)A collaborative filtering recommendation algorithm for comprehensive item scoring and attributes has been proposed.With the continuous expansion of recommender systems,the sparsity of the user-item matrix can deteriorate the performance of the traditional similarity calculation based collaborative filtering recommendation approaches and will cause cold start problem.In order to overcome this drawback,when the similarity between projects is calculated and generated,the similarity between item rate similarity and item attribute similarity is calculated by the combination of item score and item attribute,and the weight between item score and item attribute similarity is used to improve recommendation system to explore long tail project and mitigate system cold start capability.(3)The experimental results with the Movie Lens database show that the proposed algorithm can effectively alleviate the data sparsity and cold start problem faced by the traditional collaborative filtering recommendation algorithm,and improve the recommendation precision and recommendation coverage.