协同过滤技术目前被广泛应用于个性化推荐系统中.为了使用户的最近邻居集合更加精确有效,提出了基于用户兴趣度和用户特征的优化协同过滤推荐算法.首先通过计算用户对项目的兴趣度来对用户进行分组；然后采用贝叶斯算法分析出用户具有不同特征时对项目的喜好程度；最后采用一种新的相似度度量方法计算出目标用户的最近邻居集合.实验表明该算法提高了最近邻居集合的有效性和准确度,推荐质量较以往算法有明显提高.%Collaborative filtering technology is widely used in personalized recommendation system. In order to make the user' s nearest neighbors set more precise and effective, this paper presented an optimized collaborative filtering recommendation algorithm based on users' interest degree and feature. Firstly, it grouped users through calculating users' interest degree to items. Secondly, it got the value of the users' preferences for items when the users had different characteristics. Finally, it used a new method of calculating the similarity degree to calculate the target users' nearest neighbors set. The result shows that the algorithm enhances the effectiveness and accuracy of the nearest neighbors set, and the recommendation quality has significant improvement than traditional algorithm.
展开▼
