基于信息熵和用户兴趣时间性的协同过滤算法

Collaborative Filtering Algorithm Based on Information Entropy and Timeliness of User Interest

针对传统协同过滤推荐算法无法及时适应用户兴趣变化的问题,提出用信息熵来计算相似度的方法,并将两个用户对同一物品评分值的差值以及Jaccard系数加权到其中。通过在相似度计算中加入用户兴趣时间性的数据权重,提出基于信息熵和用户兴趣时间性的协同过滤算法。结果表明,该算法能够适应用户兴趣随时间变化的特性,可以有效提高推荐的精确度。

The traditional collaborative filtering recommendation algorithm can't adapt to the change of user interest in real time.This paper proposed an information entropy algorithm on similarity calculation.The difference values between two users on the same item and the Jaccard coefficient were weighted to add in the similarity calculation.By adding the data weight of interest time in similarity calculation,a collaborative filtering algorithm based on information entropy and timeliness of user interest was proposed.The results show that the algorithm can adapt to the change of user interest with time,and effectively improve the recommendation accuracy.