融合概率矩阵分解与ER规则的群组推荐方法

doi:10.3778/j.issn.1002-8331.2203-0125

摘要/Abstract

摘要： 群组推荐需要同时考虑一组内所有成员的偏好，通过融合成员偏好进而向群组推荐项目。现有的对于组推荐方法的研究中大多都将相同的权重分配给群组中所有用户，而未考虑在现实生活中不同组成员的重要性和可靠性应不同。针对该问题，提出一种新的融合概率矩阵分解与证据推理（evidence reasoning，ER）规则的群组推荐方法（FPMF-ER），以改进群组推荐中个体预测和偏好融合的过程。联合用户关系信息对经典概率矩阵分解加以改进，以获取更为完整、精准的个人预测评分；在组成员偏好融合的过程中引入ER规则，根据组成员的权重和可靠性识别群组成员的影响力，使偏好融合更为合理、准确。为了验证该方法的有效性，在Book-Crossing数据集上进行了对比实验，实验结果表明，相较于最优的基准模型，FPMF-ER的推荐结果准确性和用户满意度分别至少提高了2.55%和2.06%。

关键词: 群组推荐, 用户相关性, 概率矩阵分解, 证据推理规则

Abstract: Group recommendation needs to consider the preferences of all members in a group at the same time, and recommend items to the group by integrating member preferences. Most of the existing researches on group recommendation methods assign the same weight to all users in the group, without considering the importance and reliability of different group members in real life. To solve this problem, a group recommendation method（FPMF-ER） based on probabilistic matrix decomposition and evidence reasoning rules is proposed to improve the process of individual prediction and preference fusion in group recommendation. Firstly, the classical probability matrix decomposition is improved by combining user relationship information to obtain a more complete and accurate personal prediction score. Subsequently, ER rule is introduced in the process of preference fusion of group members, and the influence of group members is identified according to the weight and reliability of group members, so that preference fusion is more reasonable and accurate. In order to verify the effectiveness of the method, a comparative experiment is conducted on the Book-Crossing dataset. The experimental results show that compared with the optimal benchmark model, the accuracy of FPMF-ER recommendation results and user satisfaction are increased by at least 2.55% and 2.06%, respectively.

Key words: group recommendation, user relevance, probability matrix decomposition, evidence reasoning rules

王永贵, 张鉴. 融合概率矩阵分解与ER规则的群组推荐方法[J]. 计算机工程与应用, 2023, 59(5): 252-261.

WANG Yonggui, ZHANG Jian. Group Recommendation Method Integrating Probability Matrix Decomposition and ER Rules[J]. Computer Engineering and Applications, 2023, 59(5): 252-261.

参考文献

[1] LIU X C，SU X，MA J M，et al.Information filtering based on eliminating redundant diffusion and compensating balance[J].International Journal of Modern Physics B，2019，33（13）：1950129.
[2] 许晓明，梅红岩，于恒，等.基于偏好融合的群组推荐方法研究综述[J].小型微型计算机系统，2020，41（12）：2500-2508.
XU X M，MEI H Y，YU H，et al.Review of group recommendation methods based on preference fusion[J].Journal of Chinese Computer Systems，2020，41（12）：2500-2508.
[3] SINGH M，MEHROTRA M.Impact of biclustering on the performance of biclustering based collaborative filtering[J].Expert Systems with Applications，2018，113（12）：443-456.
[4] ZHANG H F，ZHOU P，JIAN Y M.Collaborative filtering recommendation algorithm based on class correlation distance[J].Recent Advances in Computer Science and Communications，2021，14（3）：887-894.
[5] 陆航，师智斌，刘忠宝.融合用户兴趣和评分差异的协同过滤推荐算法[J].计算机工程与应用，2020，56（7）：24-29.
LU H，SHI Z B，LIU Z B.Collaboration filtering recommendation algorithm based on user interest and ratings difference[J].Computer Engineering and Applications，2020，56（7）：24-29.
[6] 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 B J，YANG H J，FANG J D.Recommendation algorithm based on probability matrix factorization and fusing trust[J].Journal of Computer Applications，2019，39（10）：2834-2840.
[7] 杨丽，王时绘，朱博.基于ranking的深度张量分解群组推荐算法[J].计算机应用研究，2020，37（5）：1311-1316.
YANG L，WANG S H，ZHU B.Ranking based hybrid deep tensor factorization model for group recommendation[J].Application Research of Computers，2020，37（5）：1311-1316.
[8] 毛宇佳，刘学军，徐新艳，等.基于子组与社会行为的缩小群组推荐列表方法[J].计算机科学，2019，46（2）：202-209.
MAO Y J，LIU X J，XU X Y，et al.Method based on sub-group and social behavior for narrowing recommended list for groups[J].Computer Science，2019，46（2）：202-209.
[9] DING Z，QIN Z，WANG Q X，et al.Random group recommendation model based on fuzzy clustering[J].Journal of Electronic Science and Technology，2020，18（2）：169-178.
[10] 李晓鹏.基于模型融合的群组推荐多层集成框架[D].南京：南京邮电大学，2020.
LI X P.Multi-layer ensemble framework for group recommendation based on model fusion[D].Nanjing：Nanjing University of Posts and Telecommunications，2020.
[11] DU Y，MENG X，ZHANG Y，et al.GERF：a group event recommendation framework based on learning-to-rank[J].IEEE Transactions on Knowledge and Data Engineering，2020，32（4）：674-687.
[12] LIU Y，YIN M，ZHOU X.A collaborative filtering algorithm with intragroup divergence for poi group recommendation[J].Applied Sciences，2021，11（12）：5416.
[13] 叶柏龙，徐静静，严笋.基于评分和项目特征的群组推荐方法[J].计算机应用研究，2017，34（4）：1032-1035.
YE B L，XU J J，YAN S.Group recommendation method based on ratings and item features[J].Application Research of Computers，2017，34（4）：1032-1035.
[14] 杨金劳，刘虹明.群组信息优化矩阵分解的群组推荐方法[J].机械设计与制造，2020（4）：286-289.
YANG J L，LIU H M.Group recommendation method based on group information improved matrix decomposition[J].Machinery Design and Manufacture，2020（4）：286-289.
[15] 尹青山.基于群组信息改进矩阵分解的群组推荐方法[J].计算机应用与软件，2020，37（9）：328-333.
YIN Q S.Group recommendation method based on improved matrix decomposition of group information[J].Computer Applications and Software，2020，37（9）：328-333.
[16] GUO Z，ZENG W，WANG H，et al.An enhanced group recommender system by exploiting preference relation[J].IEEE Access，2019，7：24852-24864.
[17] WU X，WANG D，CAO W，et al.A genetic-algorithm support vector machine and D-S evidence theory based fault diagnostic model for transmission line[J].IEEE Transactions on Power Systems，2019，34（6）：4186-4194.
[18] 柯小路，马荔瑶，李子懿，等.证据推理规则的性质研究及方法修正[J].信息与控制，2016，45（2）：41-46.
KE X L，MA L Y，LI Z Y，et al.Property research and approach modification of evidential reasoning rule[J].Information and Control，2016，45（2）：41-46.
[19] GUO J，ZHU Y，LI A，et al.A social influence approach for group user modeling in group recommendation systems[J].IEEE Intelligent Systems，2016，31（5）：40-48.
[20] TANG S W，ZHOU Z J，HU C H，et al.Perturbation analysis of evidential reasoning rule[J].IEEE Transactions on Systems，Man，and Cybernetics：Systems，2019，51（8）：4895-4910.
[21] 王英博，孙永荻.基于GNN的矩阵分解推荐算法[J].计算机工程与应用，2021，57（19）：129-134.
WANG Y B，SUN Y D.GNN-based matrix factorization recommendation algorithm[J].Computer Engineering and Applications，2021，57（19）：129-134.
[22] LU Z Y，LI H，MAMOULIS N，et al.HBGG：a hierarchical Bayesian geographical model for group recommendation[C]//Proceedings of 17th SIAM International Conference on Data Mining，2017：372-380.
[23] JU C H，WANG J，XU C H.A novel application recommendation method combining social relationship and trust relationship for future internet of things[J].Multimedia Tools and Applications，2019，78（21）：29867-29880.
[24] ORTEGA F，HERNANDO A，BOBADILLA J，et al.Recommending items to group of users using matrix factorization based collaborative filtering[J].Information Sciences，2016：313-324.
[25] CASTRO J，LU J，ZHANG G，et al.Opinion dynamics-based group recommender systems[J].IEEE Transactions on Systems，Man，and Cybernetics：Systems，2018，48（12）：2394-2406.
[26] PUJAHARI A，SISODIA D S.Preference relation based collaborative filtering with graph aggregation for group recommender system[J].Applied Intelligence，2021，51（1）：1-15.
[27] 陶永才，曹朝阳，石磊，等.一种结合时间因子聚类的群组兴趣点推荐模型[J].小型微型计算机系统，2020，41（2）：356-360.
TAO Y C，CAO C Y，SHI L，et al.Group POI recommendation model based on time factor clustering[J].Journal of Chinese Computer Systems，2020，41（2）：356-360.
[28] 王永贵，林佳敏，何佳玉.融合领导者影响与隐式信任度的群组推荐方法[J].计算机工程与应用，2022，58（9）：98-106.
WANG Y G，LIN J M，HE J Y.Group recommendation method combining leader influence and implicit trust metrics[J].Computer Engineering and Applications，2022，58（9）：98-106.
[29] 吴彦文，宁彬，李斌.一种弥补群组成员差异的可解释性组推荐方法[J].小型微型计算机系统，2021，42（5）：905-977.
WU Y W，NING B，LI B.Interpretable group recommendation method to make up for differences among group members[J].Journal of Chinese Computer Systems，2021，42（5）：905-977.