数据采集与处理2016，Vol.31Issue(6)：1234-1241,8.
基于评分预测的协同过滤推荐算法
Collaborative Filtering Recommendation Algorithm Based on Rating Prediction
摘要
Abstract
T raditional collaborative filtering algorithm calculates the difference of scores only for the com‐mon items of users while calculating the similarity of users .Owing that the numbers of common items of different users is not the same ,the recommendation quality is not reliable .We proposed a new algo‐rithm ,taking both the number of common items and the popularity of goods into consideration while cal‐culating the similarity of users .Experimental results show that ,the recommendation quality of new al‐gorithm is improved by more than one time than traditional algorithm in both precision and recall .In ad‐dition ,results also show that using Pearson correlation as similarity metric obtained higher recommenda‐tion quality than Euclidean distance .关键词
推荐系统/协同过滤/相似性/召回率/准确率Key words
recommendation system/collaborative filtering (CF)/similarity/recall/precision分类
信息技术与安全科学引用本文复制引用
周海平,黄凑英,刘妮,周洪波..基于评分预测的协同过滤推荐算法[J].数据采集与处理,2016,31(6):1234-1241,8.基金项目
国家自然科学基金（11247286）资助项目；贵州省自然科学基金（黔科合J字LGK［2013］53号，黔科合LH 字［2014］7210号，黔科合 L H字［2015］7294号）资助项目；贵州省教育厅＂网络信息科学＂创新团队资助项目。 （）
