User-Based Collaborative Filtering Strategies More Interested in Improvement of Research
Yunjian Xu · Computer Technology and Development · 2011
Collaborative filtering using the similarity between users to recommend information to users interested in discovering new content,it acts as an effective technology has been applied to many fields.However,the traditional collaborative filtering algorithms can not reflect the multi-user interest and user interest changes.To address this problem,a collaborative filtering based on user clustering strategies to improve the basic idea is the basis of user-based clustering of users and more interested in that.Time threshold and the introduction of user interest data on the weight of the concepts,definitions and formulas to calculate the user preferences for different project categories and changes of interest.On this basis,they will combine the introduction of user-based collaborative filtering algorithm for clustering the recommended process.Experimental results show that the improved algorithm than the traditional collaborative filtering recommendation algorithm accuracy are dramatically increased.