基于用户评论的协同过滤推荐算法

提出融合用户评论的协同过滤推荐算法,通过挖掘电商网站的用户评论信息,获取用户评论中的产品特征和意见,通过计算每个特征意见对的极性,得到特征矩阵,结合用户意见质量形成的用户评分矩阵,求出用户评分的相似度.最后结合特征矩阵和用户评分相似度得出目标用户的综合相似度,并由预测评分得出产品推荐表,对用户进行产品推荐.实验结果表明,提出的算法与常用的推荐算法相比,改善了推荐的质量,同时推荐精度得到提高.

Collaborative Filtering Recommendation Algorithm Based on User Comments

In this paper,a new collaborative filtering recommendation algorithm based on the fusion of user comments is proposed,which is based on the mining of users' comments on the electricity supplier website to obtain the product features and related features. By using the polarity of each feature opinion pairs,the characteristic matrix is formed,and then the user rating matrix is formed by the user's opinion quality,and the similarity of the user's score is obtained. Finally,according to the characteristic matrix and the user's score similarity,the comprehensive similarity of the target user is obtained,which can predict the score and form a recommendation list. Experimental results show that the proposed algorithm improves the recommendation accuracy compared with the traditional recommendation algorithm,and improves the quality of the recommendation.