一种改进相似度的协同过滤算法实现

doi:10.16180/j.cnki.issn1007-7820.2020.02.010

摘要/Abstract

摘要：

计算相似度时,协同过滤算法会赋予所有用户或物品一致的相似度权重,进而导致相似度计算出现偏差。针对这一问题,文中提出一种改进相似度的协同过滤算法。该算法首先在计算用户间相似度时根据用户活跃量增加活跃用户惩罚因子,然后在计算物品间相似度时根据物品流行度增加热门物品惩罚因子,再对相似度做最大值归一化,最后根据相似度矩阵进行电影评分预测。实验结果表明,改进的相似度算法在评分预测时更加准确,平均绝对误差稳定在0.72左右。

关键词: 协同过滤, 皮尔逊系数, 相似度算法, 归一化, 平均绝对误差, 评分预测

Abstract:

When calculating the similarity, the collaborative filtering algorithm assigns similar weights to all users or items, which will lead to deviations in the similarity calculation. Aiming at this problem, an improved similarity algorithm was proposed to fix the error. Firstly, when calculating the similarity between users, the active user influence factor was added by the number of active users, and when calculating the similarity between items. When calculating the similarity between items, the hot item influence factor was added according to the popularity of the item, then similarity was maximum normalized. Finally, the rating of movies was predicted by using similarity matrix. The experimental results showed that the improved similarity algorithm was more accurate in rating prediction, and the average absolute error was stable at around 0.72.

Key words: collaborative filtering, pearson similarity, similarity algorithm, normalization, mean absolute error, rating prediction

中图分类号:

TP301.6

许凤翔. 一种改进相似度的协同过滤算法实现[J]. 电子科技, 2020, 33(2): 54-59.

XU Fengxiang. Implementation of a Collaborative Filtering Algorithm Based on Improved Similarity[J]. Electronic Science and Technology, 2020, 33(2): 54-59.

图/表 4

参考文献 16

[1]	高良友, 黄梦醒 . 综合用户特征和项目属性的协作过滤推荐算法改进[J]. 海南大学学报(自然科学版), 2015,33(2):135-140.
	Gao Liangyou, Huang Mengxing . Improved collaborative filtering recommendation algorithm for integrated user features and project attributes[J]. Journal of Hainan University(Natural Science Edition), 2015,33(2):135-140.
[2]	王晓军 . 利用项目属性和偏好改进协同过滤推荐[J]. 北京邮电大学学报, 2014,37(6):68-71.
	Wang Xiaojun . Employing item attribute and preference to enhance the collaborative filtering recommendation[J]. Journal of Beijing University of Posts and Telecommunications, 2014,37(6):68-71.
[3]	于洪, 李俊华 . 一种解决新项目冷启动问题的推荐算法[J]. 软件学报, 2015,26(6):1395-1408.
	Yu Hong, Li Junhua . Algorithm to solve the cold-start problem in new item recommendations[J]. Journal of Software, 2015,26(6):1395-1408.
[4]	王建芳, 刘冉东, 谷振鹏 , 等. 一种改进专家信任的协同过滤推荐算法[J]. 计算机应用研究, 2018(2):354-357.
	Wang Jianfang, Liu Randong, Gu Zhenpeng , et al. Improved expert trust collaborative filtering recommendation algorithm[J]. Application Research of Computers, 2018(2):354-357
[5]	硕良勋, 柴变芳, 张新东 . 基于改进最近邻的协同过滤推荐算法[J]. 计算机工程与应用, 2015,51(5):137-141.
	Shuo Liangxun, Chai Bianfang, Zhang Xingdong . Collaborative filtering algorithm based on improved nearest neighbors[J]. Computer Engineering and Applications, 2015,51(5):137-141.
[6]	焦明海, 陈晓芳, 陈旭 , 等. 基于贝叶斯网络认知反馈的协同过滤推荐[J]. 控制工程, 2017,24(7):1310-1317.
	Jiao Minghai, Chen Xiaofang, Chen Xu , et al. Collaborative filtering based on bayesian network and cognitive feedback[J]. Control Engineering of China, 2017,24(7):1310-1317.
[7]	肖宇航, 吴明礼 . 改进用户相似度的协同过滤算法应用研究[J]. 信息技术, 2018(7):138-142.
	Xiao Yuhang, Wu Mingli . The application and research on advanced user similarity on user-based collaborative filtering[J]. Information Technology, 2018(7):138-142.
[8]	肖文强, 姚世军, 吴善明 . 一种改进的Top-N协同过滤推荐算法[J]. 计算机应用研究, 2018,35(1):105-108.
	Xiao Wenqiang, Yao Shijun, Wu Shanming . Improved Top-N collaborative filtering recommend algorithm[J]. Application Research of Computers, 2018,35(1):105-108.
[9]	王余斌, 王成良 . 基于用户评论评分与信任度的协同过滤算法[J]. 计算机应用研究, 2018,35(5):1368-1371.
	Wang Yubin, Wang Chengliang, Wen Junhao . Research on collaborative filtering recommendation algorithm based on ratings reviews and user trust[J]. Application Research of Computers, 2018,35(5):1368-1371.
[10]	李熠晨, 陈莉, 石晨晨 , 等. 采用信任网络增强的协同过滤算法[J]. 计算机应用研究, 2018,35(1):116-120.
	Li Yichen, Chen Li, Shi Chenchen , et al. Enhanced collaborative filtering adopting trust network[J]. Application Research of Computers, 2018,35(1):116-120.
[11]	王宝林, 韩帅帅, 张德海 . 一种基于社会化标签的协同过滤推荐算法[J]. 电子科技, 2015,28(7):90-93.
	Wang Baolin, Han Shuaishuai, Zhang Dehai . Collaborative filtering recommendation algorithm based on social tags[J]. Electronic Science and Technology, 2015,28(7):90-93.
[12]	张雅科 . 基于模糊权重相似性的协同过滤算法研究[J]. 电子科技, 2015,28(7):111-114.
	Zhang Yake . A study of fuzzy weighted similarity measure for collaborative filtering recommender systems[J]. Electronic Science and Technology, 2015,28(7):111-114.
[13]	Petroni F, Querzoni L, Beraldi R , et al. LCBM:a fast and lightweight collaborative filtering algorithm for binary ratings[J]. Journal of Systems & Software, 2016,117(6):583-594.
[14]	李欢 . 新型协同过滤推荐算法研究[D]. 合肥:安徽大学, 2017.
	Li Huan . Research on a new collaborative filtering recommendation algorithm[D]. Hefei:Anhui University, 2017.
[15]	肖宇航 . 行为分析技术在用户管理系统中的应用研究[D]. 北京:北方工业大学, 2018.
	Xiao Yuhang . Application and research of behavioral analysis technology in user management system[D]. Beijing:North China University of Technology, 2018
[16]	李玉省 . 个性化推荐系统关键技术研究[D]. 北京:北京邮电大学, 2016.
	Li Yusheng . Research on some key technologies of personalization recommendation system[D]. Beijing:Beijing University of Posts and Telecommunications, 2016.

相似度	优点	缺点
余弦相似度	适用于文本计算	对具体数值不敏感
欧几里得相似度	适用于空间距离映射	只关注于数值,未考虑评分尺度
皮尔逊相似度	进行了共同评分去中心化	用户矩阵稀疏时无法计算用户相似度
修正余弦相似度	进行了所有评分去中心化	对用户未评分项目假设为0