Collaborative Filtering Recommendation Based On User Clustering For Item's Multiple Contents

In recent years, with the popularity of the Internet and e-commerce to flourish, e-commerce personalization recommendation systems, especially collaborative filtering recommender systems, have gradually become an important e-commerce research. The tremendous growth in the amount of available information and the kinds of commodities to web sites poses some key challenges for recommender systems, so the problems of cold-start and sparsity in collaborative filtering recommendation are needed to be solved.Collaborative filtering recommendation algorithm can make choices based on the opinions of other people. It is the most successful technology for building recommender systems to date. Unfortunately, traditional collaborative filtering algorithm does not consider the problem of item's multiple contents and often leads to bad recommendation when item has multiple contents. To solve this problem, an organic combination of the recommendation algorithm, which based on users and recommendation algorithm based on items and clustering algorithm is proposed, which can greatly improve the issues, which are cold-start, sparse, real-time. First, obtain similar items by item-based collaborative filtering algorithm and predict users'initial ratings of items which these users did not rate by using similarity of similar items, in order to fill the user - item rating matrix. Second, on the scope of the similar items, with using clustering technology and user-based collaborative filtering algorithm, predict the active user's final rating scores for items and come to the recommended list.Finally, from the relevant experiments and analyses, the collaborative filtering recommendation based on user clustering for item's multiple contents is more effective and advanced than tradition technology in quality of recommendation; it can meet users'demands better.