融合信任和基于概率矩阵分解的推荐算法

doi:10.11772/j.issn.1001-9081.2019030583

计算机应用 ›› 2019, Vol. 39 ›› Issue (10): 2834-2840.DOI: 10.11772/j.issn.1001-9081.2019030583

融合信任和基于概率矩阵分解的推荐算法

田保军¹, 杨浒昀², 房建东¹

1. 内蒙古工业大学信息工程学院, 呼和浩特 010080;
2. 内蒙古工业大学数据科学与应用学院, 呼和浩特 010080

收稿日期:2019-04-09 修回日期:2019-05-22 发布日期:2019-05-31 出版日期:2019-10-10
通讯作者: 田保军
作者简介:田保军(1971-),男,内蒙古呼和浩特人,副教授,硕士,主要研究方向:机器学习、推荐系统;杨浒昀(1993-),男,内蒙古巴彦淖尔人,硕士研究生,主要研究方向:推荐系统;房建东(1966-),女,内蒙古呼和浩特人,教授,博士,主要研究方向:信息处理、智能控制。
基金资助:
内蒙古自治区自然科学基金资助项目（2015MS0613）；内蒙古自治区科技重大项目（2018ZD0302）；内蒙古自治区科技计划项目（20170306）。

Recommendation algorithm based on probability matrix factorization and fusing trust

TIAN Baojun¹, YANG Huyun², FANG Jiandong¹

1. College of Information Engineering, Inner Mongolia University of Technology, Hohhot Nei Mongol 010080, China;
2. College of Data Science and Application, Inner Mongolia University of Technology, Hohhot Nei Mongol 010080, China

Received:2019-04-09 Revised:2019-05-22 Online:2019-05-31 Published:2019-10-10
Supported by:
This work is partially supported by the Inner Mongolia Autonomous Region Natural Science Fund Project of China (2015MS0613), the Inner Mongolia Autonomous Region Science and Technology Major Project of China (2018ZD0302), the Inner Mongolia Autonomous Region Science and Technology Plan Project of China (20170306).

摘要/Abstract

摘要： 针对推荐精度不准确、数据稀疏、恶意推荐的问题，提出融合信任基于概率矩阵分解（PMF）的新推荐模型。首先，通过建立基于信任的协同过滤模型（CFMTS）将改进的信任机制融入到协同过滤推荐算法中。信任值通过全局信任及局部信任计算获得，其中局部信任利用了信任传播机制计算用户的直接信任值和间接信任值得到，全局信任采用信任有向图的方式计算得到。然后，将信任值与评分相似度融合以解决数据稀疏、恶意推荐的问题。同时，将CFMTS融入到PMF模型中以建立新的推荐模型——融合信任基于概率矩阵分解模型（MPMFFT），通过梯度下降算法对用户特征向量和项目特征向量进行计算以产生预测评分值，进一步提高推荐系统的精准度。通过实验将提出的MPMFFT与经典的PMF、社交信息的矩阵分解（SocialMF）、社交信息的推荐（SoRec）、加权社交信息的推荐（RSTE）等模型进行了结果的对比和分析，在公开的真实数据集Epinions上MPMFFT的平均绝对误差（MAE）和均方根误差（RMSE）比最优的RSTE模型分别降低2.9%和1.5%，同时在公开的真实数据集Ciao上MPMFFT的MAE和RMSE比最优的SocialMF模型分别降低1.1%和1.8%，结果证实了模型能在一定程度上解决数据稀疏、恶意推荐问题，有效提高推荐质量。

关键词: 推荐系统, 信任关系, 概率矩阵分解, 特征向量

Abstract: For the problems of low recommendation accuracy, data sparsity and malicious recommendation, a new recommendation model based on Probability Matrix Factorization (PMF) and fusing trust was proposed. Firstly, by establishing a Collaborative Filtering Model based on Trust Similarity (CFMTS), the improved trust mechanism was integrated into the collaborative filtering recommendation algorithm. The trust value was obtained through global trust and local trust calculation. The local trust was obtained by calculating the direct trust value and the indirect trust value of the user by the trust propagation mechanism, the global trust was calculated by the trust directed graph. Then, the trust value was combined with the score similarity to solve the problems of data sparsity and malicious recommendation. At the same time, CFMTS was integrated into the PMF model to establish a new recommendation model-Model based on Probability Matrix Factorization and Fusing Trust (MPMFFT). The user feature vectors and the project feature vectors were calculated by the gradient descent algorithm to generate the predicted scores, further improving the accuracy of the recommender system. Through experiments, the proposed MPMFFT was compared with the classical models such as PMF, Social Matrix Factorization (SocialMF), Social Recommendation (SoRec) and Recommendations with Social Trust Ensemble (RSTE). The proposed model has the Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) decreased by 2.9% and 1.5% respectively compared with the optimal model RSTE on the open real dataset Epinions, and has the MAE and RMSE decreased by 1.1% and 1.8% respectively compared with the optimal SocialMF model on open real dataset Ciao, verifying that the proposed model is significantly improved on the above indicators. The results confirme that the propose model can resolve the problem of data sparseness and malicious recommendation to some extent, and effectively improved the recommendation quality.

Key words: recommender system, trust relationship, Probability Matrix Factorization (PMF), feature vector

中图分类号:

TP183

田保军, 杨浒昀, 房建东. 融合信任和基于概率矩阵分解的推荐算法[J]. 计算机应用, 2019, 39(10): 2834-2840.

TIAN Baojun, YANG Huyun, FANG Jiandong. Recommendation algorithm based on probability matrix factorization and fusing trust[J]. Journal of Computer Applications, 2019, 39(10): 2834-2840.

参考文献

[1] 中国互联网络信息中心. 第43次中国互联网络发展状况统计报告[EB/OL].[2019-03-12]. http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201902/P020190318523029756345.pdf.(China Internet Network Information Center. The 43rd China Statistical Report on Internet Development[EB/OL].[2019-03-12]. http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201902/P020190318523029756345.pdf.)
[2] WANG P. A collaborative filtering recommendation algorithm based on product clustering[J]. Applied Mechanics & Materials, 2013, 267:87-90.
[3] AZADJALAL M M, MORADI P, ABDOLLAHPOURI A, et al. A trust-aware recommendation method based on Pareto dominance and confidence concepts[J]. Knowledge-Based Systems, 2017, 116:130-143.
[4] SHU J, SHEN X, LIU H, et al. A content-based recommendation algorithm for learning resources[J]. Multimedia Systems, 2017, 24(2):163-173.
[5] 王余斌, 王成良, 文俊浩. 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 Y B, WANG C L, WEN J H. Research on collaborative filtering recommendation algorithm based on ratings, reviews and user trust[J]. Application Research of Computers, 2018, 35(5):1368-1371.)
[6] LIU H, KONG X, BAI X, et al. Context-based collaborative filtering for citation recommendation[J]. IEEE Access, 2015, 3:1695-1703.
[7] MATHEW P, KURIAKOSE B, HEGDE V. Book recommendation system through content based and collaborative filtering method[C]//Proceedings of the 2016 International Conference on Data Mining and Advanced Computing. Piscataway:IEEE, 2016:47-52.
[8] LI W, JIAN C, WU J, et al. A collaborative filtering recommendation method based on discrete quantum-inspired shuffled frog leaping algorithms in social networks[J]. Future Generation Computer Systems, 2018, 88:262-270.
[9] 于波, 陈庚午, 王爱玲, 等. 一种结合项目属性的混合推荐算法[J]. 计算机系统应用, 2017, 26(1):147-151. (YU B, CHEN G W, WANG A L, et al. Hybrid recommendation algorithm combined with the project properties[J]. Computer Systems & Applications, 2017, 26(1):147-151.)
[10] 杨丰瑞, 郑云俊, 张昌. 结合概率矩阵分解的混合型推荐算法[J]. 计算机应用, 2018, 38(3):644-649. (YANG F R, ZHENG Y J, ZHANG C. Hybrid recommendation algorithm based on probability matrix factorization[J]. Journal of Computer Applications, 2018, 38(3):644-649.)
[11] MENG X. speedTrust:a super peer-guaranteed trust model in hybrid P2P networks[J]. Journal of Supercomputing, 2018, 74(6):2553-2580.
[12] MASSA P, BHATTACHARJEE B. Using trust in recommender systems:an experiment alanalysis[C]//Proceedings of the 2004 International Conference on Trust Management, LNCS 2995. Berlin:Springer, 2004:221-235.
[13] GOLBECK J, GERHARD J, O'COLMAN F, et al. Scaling up integrated structural and content-based network analysis[J]. Information Systems Frontiers, 2018, 20(6):1191-1202.
[14] MASSA P, AVESANI P. Trust-aware recommender systems[C]//Proceedings of the 2007 ACM Conference on Recommender Systems. New York:ACM, 2007:17-24.
[15] JAMALI M, ESTER M. Trustwalker:a random walk model for combining trust-based and itembasedrecommendation[C]//Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM, 2009:397-406.
[16] GUO G, ZHANG J, ZHU F, et al. Factored similarity models with social trust for top-N item recommendation[J]. Knowledge-Based Systems, 2017, 122:17-25.
[17] ZHANG T, LI W, WANG L, et al. Social recommendation algorithm based on stochastic gradient matrix decomposition in social network[EB/OL].[2019-01-10]. https://doi.org/10.1007/s12652-018-1167-7.
[18] NAZEMIAN A, GHOLAMI H, TAGHIYAREH F. An improved model of trust-aware recommender systems using distrust metric[C]//Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining. Piscataway:IEEE, 2012:1079-1084.
[19] UZZI B, SPIRO J. Collaboration and creativity:the small world problem[J]. American Journal of Sociology, 2005, 111(2):447-504.

融合信任和基于概率矩阵分解的推荐算法

Recommendation algorithm based on probability matrix factorization and fusing trust

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	刘欢, 李晓戈, 胡立坤, 胡飞雄, 王鹏华. 基于知识图谱驱动的图神经网络推荐模型[J]. 计算机应用, 2021, 41(7): 1865-1870.
[2]	林怿星, 唐华. 基于异构信息网络的混合推荐模型[J]. 计算机应用, 2021, 41(5): 1348-1355.
[3]	高铭蔚, 桑楠, 杨茂林. 基于胶囊网络的交互式网络电视视频点播推荐模型[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3171-3177.
[4]	任柯舟, 彭甫镕, 郭鑫, 王喆, 张晓静. 动态融合社交信息的社会化推荐[J]. 计算机应用, 2021, 41(10): 2806-2812.
[5]	李翔锟, 贾彩燕. 融合重叠社区正则化及隐式反馈的协同过滤方法[J]. 计算机应用, 2021, 41(1): 53-59.
[6]	李春英, 汤庸, 肖政宏, 李天送. 学术社交网络中的权威学者推荐模型[J]. 计算机应用, 2020, 40(9): 2594-2599.
[7]	田保军, 刘爽, 房建东. 融合主题信息和卷积神经网络的混合推荐算法[J]. 计算机应用, 2020, 40(7): 1901-1907.
[8]	魏哲, 王小华. 淋巴结转移检测的八度卷积方法[J]. 计算机应用, 2020, 40(3): 723-727.
[9]	秦婧, 张青博, 王斌. 关注长尾物品的推荐方法[J]. 计算机应用, 2020, 40(2): 454-458.
[10]	张文龙, 钱付兰, 陈洁, 赵姝, 张燕平. 基于双重最相关注意力网络的协同过滤推荐算法[J]. 计算机应用, 2020, 40(12): 3445-3450.
[11]	崔艺馨, 陈晓东. Spark框架优化的大规模谱聚类并行算法[J]. 计算机应用, 2020, 40(1): 168-172.
[12]	张延良, 卢冰, 洪晓鹏, 赵国英, 张伟涛. 基于局部区域方法的微表情识别[J]. 计算机应用, 2019, 39(5): 1282-1287.
[13]	邵长城, 陈平华. 融合社交网络和图像内容的兴趣点推荐[J]. 计算机应用, 2019, 39(5): 1261-1268.
[14]	王磊, 任航, 龚凯. 基于多维信任和联合矩阵分解的社会化推荐方法[J]. 计算机应用, 2019, 39(5): 1269-1274.
[15]	雷曼, 龚琴, 王纪超, 王保群. 基于标签权重的协同过滤推荐算法[J]. 计算机应用, 2019, 39(3): 634-638.