摘要
为提高推荐算法的精度,提出一种基于从众心理矩阵和加权相似度的推荐算法。引入项目流行度阈值K,根据项目流行度对项目进行筛选,对高流行度项目的评分矩阵进行调整;通过用户兴趣度量函数度量用户的现阶段偏好项目,基于每个用户的现阶段偏好项目计算用户间的兴趣相似度;采用相似度加权融合的方式获取用户相似度。实验结果表明,该算法的推荐精度优于传统的协同过滤算法。
To improve the accuracy of recommendation algorithm,a recommendation algorithm based on crowd psychology matrix and weighted similarity was proposed.The project popularity threshold K was introduced,and the project was screened according to the popularity of the project,the score of the high prevalence project was adjusted.The user's current preference items were measured using the user interest metric function,and the interest similarity between users was calculated based on the current phase preference of each user.Similarity weighted fusion was adopted to get user similarity.Experimental results show that the proposed algorithm is superior to the traditional collaborative filtering algorithm in terms of recommendation accuracy.
引文
[1]ZHANG Xinmeng,LI Song.Collaborative filtering recommendation algorithm based on item attribute and data weight[J].Automation and Instrumentation,2016,31(9):69-73(in Chinese).[张新猛,李松.基于项目属性与数据权重的协同过滤推荐算法[J].自动化与仪表,2016,31(9):69-73.]
[2]WENG Xiaolan,WANG Zhijian.Progress in collaborative filtering recommendation algorithm[J].Computer Engineering and Applications,2018,54(1):25-31(in Chinese).[翁小兰,王志坚.协同过滤推荐算法研究进展[J].计算机工程与应用,2018,54(1):25-31.]
[3]HAN Yanan,CAO Han,LIU Liangliang,et al.Collaborative filtering recommendation algorithm based on score matrix filling and user interest[J].Computer Engineering,2016,42(1):36-40(in Chinese).[韩亚楠,曹菡,刘亮亮,等.基于评分矩阵填充与用户兴趣的协同过滤推荐算法[J].计算机工程,2016,42(1):36-40.]
[4]GAO Changyuan,HUANG Kai,WANG Jing,et al.Collaborative filtering recommendation algorithm based on commodity attribute value and user characteristics[J].Computer Engineering and Science,2017,39(12):2333-2339(in Chinese).[高长元,黄凯,王京,等.基于商品属性值和用户特征的协同过滤推荐算法[J].计算机工程与科学,2017,39(12):2333-2339.]
[5]YU Jinming,MENG Jun,WU Qiufeng.Project collaborative filtering recommendation algorithm based on improved similarity measure[J].Computer Application,2017,37(5):1387-1391(in Chinese).[于金明,孟军,吴秋峰.基于改进相似性度量的项目协同过滤推荐算法[J].计算机应用,2017,37(5):1387-1391.]
[6]LIU Liangliang,CAO Han,HAN Yanan.Collaborative filtering algorithm based on group dynamics and its application[J].Computer Application Research,2014,31(12):3603-3605(in Chinese).[刘亮亮,曹菡,韩亚楠.基于群体动力学的协同过滤算法及应用[J].计算机应用研究,2014,31(12):3603-3605.]
[7]WANG Wei,ZHENG Jun.Improvement of collaborative filtering algorithm based on user similarity[J].Journal of East China Normal University,2016,5(3):61-66(in Chinese).[王威,郑骏.基于用户相似度的协同过滤算法改进[J].华东师范大学学报,2016,5(3):60-66.]
[8]RONG Huijia,HUO Shengxu,HU Chunhua,et al.Collaborative filtering recommendation algorithm based on user similarity[J].Journal of Communication,2014,35(2):17-24(in Chinese).[荣辉佳,火生旭,胡春华,等.基于用户相似度的协同过滤推荐算法[J].通信学报,2014,35(2):17-24.]
[9]SUN Hui,MA Yue,YANG Meibo,et al.An improved similarity based user clustering collaborative filtering recommendation algorithm[J].Small Microcomputer System,2014,35(9):1967-1970(in Chinese).[孙辉,马跃,杨梅波,等.一种相似度改进的用户聚类协同过滤推荐算法[J].小型微型计算机系统,2014,35(9):1967-1970.]
[10]FENG Lili.Relationship between herd mentality and consumption behavior[J].Chinese and Foreign Entrepreneurs,2013(7):214(in Chinese).[封梨梨.从众心理与消费行为关系[J].中外企业家,2013(7):214.]
[11]NIU Kangkang.Analysis of the influence mechanism of popularity on user interest and its application in recommendation algorithm[D].Shanghai:Shanghai University,2015(in Chinese).[牛抗抗.流行度对用户兴趣的影响机制分析及其在推荐算法中的应用研究[D].上海:上海大学,2015.]
[12]Fan X,Hu Y,Zhang R,et al.Modeling temporal effectiveness for context-aware web services recommendation[C]//IEEE International Conference on Web Services.IEEE,2015:225-232.
[13]YANG Li,HU Yunhong,SHAO Guirong.A method for acquiring collaborative preferences integrating time decay and preference fluctuation[J].Computer Applications,2016,36(7):2011-2015(in Chinese).[杨立,胡运红,邵桂荣.融合时间衰减与偏好波动的协同偏好获取方法[J].计算机应用,2016,36(7):2011-2015.]