一种基于用户特征和时间的协同过滤算法

A Collaborative Filtering Recommendation Based on User Characteristics and Time Weight

协同过滤是个性化推荐系统中采用最广泛的推荐技术,但已有的方法是将用户不同时间的兴趣等同考虑,时效性不足,而且相同用户特征的用户兴趣存在着很大的相似性,针对此问题,提出一种基于用户特征和时间的协同过滤算法,使得越接近采集时间的用户兴趣,在推荐过程中具有更大的权值,并且根据用户的特征来来提高相似用户集的采集,从而提高推荐的准确性。

Collaborative filtering is the most widely used recommendation technology in the personalized recommendation system. However the user＇s interests in different time have been taken into equal consideration with the method being used, which leads to the lack of effectiveness in the given period of time. The interests of the users who have the same characteristics are very similar. In view of this problem, this paper presents an improved collaborative filtering algorithm, which based on user characteristics and time weight, to make the click interests approaching the gathering time, make the weight of recommendation process bigger, and according to the characteristics of users to enhance the acquisition of similar user, thereby to improve the accuracy of the recommendation.