【摘要】 传统协同过滤推荐算法依据共同评分项目计算用户相似度,进而产生推荐项目。然而,随着用户和商品数量的不断增加,用户共同评分的项目会越来越少,甚至没有,因此传统协同过滤推荐算法对用户之间相似度的衡量将会越来越不准确,从而影响推荐系统的性能。针对这一问题,本文对用户相似度的计算方法进行了改进,提出直接相似度和间接相似度的概念,同时引入关键人物权重,进一步提高推荐系统的准确性。更多还原

【Abstract】 In traditional collaborative filtering recommendation Algorithm,similarity of users is often calculated based on common ratings,and then the recommended items are produced. However,with the increasing number of users and products,the common rated items will be less and less,and even no. So the measure of the similarity of users will be more and more inaccurate,and thus it will affect the performance of the recommendation system. In order to solve this problem,the method of calculating the similarity of users is improved,and the concepts of direct similarity and indirect similarity are put forward. At the same time,in order to further improve the accuracy of the recommendation system,the key figure is introduced into the system.更多还原