摘要
引入云模型改进基于标签的用户相似性和资源相似性度量方法,进而提出了基于标签和云模型的协同过滤算法.通过在Movie Lens数据集上的实验表明:改进后的算法在precision,recall和F1-measure三个指标上均取得较好的推荐效果,推荐效率均优于传统的方法.
In order to use the tag information more accurately reflect the characteristics of users and resources, the cloud model was introduced to improve the user similarity and resource similarity measurement method based on tags,and a collaborative filtering algorithm based on tag and cloud model was proposed. Experiments on the MovieLens data set showed that the improved algorithm has better recommendation effects on three evaluation metrics which are precision, recall and FI-measure, the effectiveness of the recommendation is better than traditional method.
出处
《中南民族大学学报(自然科学版)》
CAS
北大核心
2016年第3期117-122,共6页
Journal of South-Central University for Nationalities:Natural Science Edition
基金
湖北省自然科学基金资助项目(2013CFB445)
中南民族大学研究生创新基金资助项目(2016sycxjj207)
关键词
标签
协同过滤
云模型
相似性
tag
collaborative filtering
cloud m odel
similarity