基于用户属性和项目类别的协同过滤算法

针对传统的协同过滤算法中数据稀疏性所导致的推荐系统推荐质量不高的问题,文章结合用户和产品背景信息,对其进行加权处理,提出了基于用户和产品信息加权的协同过滤算法.该方法首先计算基于用户属性的相似性和基于项目类别的相似性,然后将计算的结果作为加权值融合到传统的相似度计算中,弥补因为数据稀疏而造成不能准确地进行个性化推荐的不足,提供更多可参考数据进行精确推荐.实验结果表明,该算法能有效提高推荐质量,产生较好的推荐效果.

Collaborative Filtering Algorithm Based on User Property and Item Category

Aiming at the defect reduced by the data sparsity in the collaborative filtering used in common currently,a collaborative filtering algorithm based on users’ and items’ information is proposed,which utilized users’ and items’ background information to advanced the conventional collaborative filtering by weighting them.This method calculate the similarity based on user property and the similarity based on product category,and then adopted the results into the conventional one to offset the deficiency of inappropriate personalized recommendation due to the lack of background information and provide more useful information to recommend precisely,which proved to be more efficient and accurate in the research.