Adaptive social recommendation based on negative similarity

doi:10.11772/j.issn.1001-9081.2022071003

Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (8): 2439-2447.DOI: 10.11772/j.issn.1001-9081.2022071003

Special Issue: 数据科学与技术

• Data science and technology • Previous Articles Next Articles

Adaptive social recommendation based on negative similarity

Yinying ZHOU¹, Yunsheng ZHOU¹, Dunhui YU¹^,², Jun SUN¹()

^1.School of Computer Science and Information Engineering，Hubei University，Wuhan Hubei 430062，China
^2.Hubei Provincial Engineering and Technology Research Center for Education Informatization （Hubei University），Wuhan Hubei 430062，China

Received:2022-07-13 Revised:2022-11-06 Accepted:2022-11-11 Online:2023-01-15 Published:2023-08-10
Contact: Jun SUN
About author:ZHOU Yinying， born in 1998， M. S. candidate. Her research interests include social recommendation.
ZHOU Yunsheng， born in 2002. His research interests include crowdsourcing.
YU Dunhui， born in 1974， Ph. D.， professor. His research interests include knowledge graph， crowdsourced data management， service computing.
Supported by:
National Scientific and Technological Innovation 2030 - Major Project(2020AAA0107700);National Natural Science Foundation of China(61977021)

基于消极相似性的自适应社会化推荐

周寅莹¹, 周允升¹, 余敦辉¹^,², 孙军¹()

^1.湖北大学计算机与信息工程学院，武汉 430062
^2.湖北省教育信息化工程技术研究中心（湖北大学），武汉 430062

通讯作者: 孙军
作者简介:周寅莹（1998—），女，湖北随州人，硕士研究生，CCF会员，主要研究方向：社会化推荐
周允升（2002—），男，湖北恩施人，主要研究方向：众包
余敦辉（1974—），男，湖北武汉人，教授，博士，CCF会员，主要研究方向：知识图谱、众包数据管理、服务计算；
基金资助:
国家科技创新2030—重大项目(2020AAA0107700);国家自然科学基金资助项目(61977021)

Abstract

Abstract:

Social recommendation aims to improve recommendation effect of traditional recommendation algorithms by integrating social relations. Currently， social recommendation algorithms based on Network Embedding （NE） face two problems： one is that inconsistency between objects is not considered when constructing network， and algorithms are often restricted by positive objects that are difficult to obtain and have many constraints； the other is that the elimination of overfitting in algorithm training process based on the number of ratings cannot be realized by these algorithms. Therefore， an Adaptive Social Recommendation algorithm based on Negative Similarity （ASRNS） was proposed. Firstly， homogeneous networks with positive correlations were constructed by consistency analysis. Then， embedded vectors were obtained by combining weighted random walk with Skip-Gram algorithm. Next， similarities were calculated， and Matrix Factorization （MF） algorithm was constrained from the perspective of negative similarity. Finally， the number of ratings was mapped to the ideal rating range based on adaptive mechanism， and different penalties were imposed on bias terms of the algorithm. Experiments were conducted on FilmTrust and CiaoDVD datasets. The results show that compared with algorithms such as Collaborative User Network Embedding （CUNE） algorithm and Consistent neighbor aggregation for Recommendation （ConsisRec） algorithm， ASRNS has the Root Mean Square Error （RMSE） reduced by at least 2.60% and 5.53% respectively， and the Mean Absolute Error （MAE） reduced by at least 1.47% and 2.46% respectively. It can be seen that ASRNS can not only reduce rating prediction error effectively， but also improve over-fitting problem in algorithm training process significantly， and has good robustness for objects with different ratings.

Key words: social recommendation, network embedding, negative similarity, adaptive mechanism, Matrix Factorization (MF)

摘要：

社会化推荐旨在融合社会关系改善传统推荐算法的推荐效果。当前基于网络嵌入（NE）的社会化推荐算法面临两个问题：一是在构建网络时未考虑对象间的不一致性，并且倾向于利用获取难度大、约束条件多的积极对象来约束算法；二是这些算法未能依据评分数量消除算法训练中的过拟合。因此，提出一种基于消极相似性的自适应社会化推荐（ASRNS）算法。首先通过一致性分析构建具有正向相关性的同构网络；接着联合加权随机游走与Skip-Gram算法得到嵌入向量；然后计算相似度，并从消极相似性的角度来约束矩阵分解（MF）算法；最后基于自适应机制将评分数量映射到理想评分数量区间，并对算法偏置项施加不同的惩罚。在FilmTrust和CiaoDVD数据集上实验结果表明，与协同用户网络嵌入（CUNE）算法、一致性邻居聚合的推荐（ConsisRec）算法等算法相比，ASRNS的均方根误差（RMSE）分别至少降低了2.60%和5.53%，平均绝对误差（MAE）分别至少降低了1.47%和2.46%。可见，ASRNS不仅可以有效降低评分预测误差，还能显著改善算法训练过程中的过拟合问题，对不同评分数量的对象都具有较好的健壮性。

CLC Number:

TP391.3

Yinying ZHOU, Yunsheng ZHOU, Dunhui YU, Jun SUN. Adaptive social recommendation based on negative similarity[J]. Journal of Computer Applications, 2023, 43(8): 2439-2447.

周寅莹, 周允升, 余敦辉, 孙军. 基于消极相似性的自适应社会化推荐[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2439-2447.

Figures/Tables 17

Fig. 1 Overall framework of ASRNS

Tab. 1 Examples of consistency calculation

用户对	$C u v$	$H u v$	$A u v$
$u 1, u 5$	$i 1, i 2, i 4, i 6$	$i 2, i 4, i 6$	$i 1$
$u 2, u 3$	$i 1, i 5, i 6$	$i 1$	$∅$
$u 2, u 4$	$i 2, i 4, i 5$	$∅$	$i 4$

Tab. 1 Examples of consistency calculation

用户对	$C u v$	$H u v$	$A u v$
$u 1, u 5$	$i 1, i 2, i 4, i 6$	$i 2, i 4, i 6$	$i 1$
$u 2, u 3$	$i 1, i 5, i 6$	$i 1$	$∅$
$u 2, u 4$	$i 2, i 4, i 5$	$∅$	$i 4$

Tab. 2 Statistics of experimental datasets

数据集	用户数	项目数	评分数	关系数
FilmTrust	1 508	2 071	35 497	1 853
CiaoDVD	17 615	16 121	72 665	40 133

Tab. 3 Experimental parameter setting

参数	取值范围
一致性程度 $θ$	0.2， 0.3， 0.4， 0.5， 0.6， 0.7
权重 $ω$	1.0， 0.9， 0.8， 0.7， 0.6， 0.5， 0
消极对象数 $y$	10， 20， 30， 40， 50， 60
理想评分数量区间 $[D s, D d]$	［11，20］，［21，30］，［31，40］，［41，50］，［51，60］，［61，70］

Tab. 3 Experimental parameter setting

参数	取值范围
一致性程度 $θ$	0.2， 0.3， 0.4， 0.5， 0.6， 0.7
权重 $ω$	1.0， 0.9， 0.8， 0.7， 0.6， 0.5， 0
消极对象数 $y$	10， 20， 30， 40， 50， 60
理想评分数量区间 $[D s, D d]$	［11，20］，［21，30］，［31，40］，［41，50］，［51，60］，［61，70］

Fig. 2 Influence of θ on evaluation indicators

Fig. 3 Influence of ω on evaluation indicators

Fig. 4 Influence of y on evaluation indicators

Fig. 5 Influence of [Ds,Dd] on evaluation indicators

Tab. 4 Comparison of prediction results

数据集	评价指标	ASRNS-C	ASRNS	降低比例/%
F（全体）	RMSE	0.801	0.786	1.87
F（全体）	MAE	0.615	0.605	1.63
F（冷启动）	RMSE	0.871	0.857	1.61
F（冷启动）	MAE	0.692	0.682	1.45
C（全体）	RMSE	0.911	0.907	0.44
C（全体）	MAE	0.669	0.666	0.45
C（冷启动）	RMSE	0.933	0.922	1.18
C（冷启动）	MAE	0.681	0.674	1.03

Fig. 6 Comparison of convergence of loss function

Tab. 5 Comparison of experimental results of MF-type algorithms on all datasets

算法	FilmTrust				CiaoDVD
	k=5		k=10		k=5		k=10
	RMSE	MAE	RMSE	MAE	RMSE	MAE	RMSE	MAE
LOCABAL	0.839	0.651	0.830	0.643	1.028	0.778	1.023	0.781
SREPS	0.867	0.663	0.849	0.651	1.033	0.782	1.031	0.784
HERec	0.835	0.642	0.829	0.635	1.027	0.777	1.021	0.780
SREE	0.815	0.621	0.811	0.620	1.005	0.752	1.001	0.757
CUNE	0.825	0.631	0.820	0.627	0.988	0.727	0.981	0.722
ASRNS	0.790	0.608	0.786	0.605	0.908	0.676	0.906	0.675

Tab. 6 Comparison of experimental results of MF-type algorithms on cold start datasets

算法	FilmTrust				CiaoDVD
	k=5		k=10		k=5		k=10
	RMSE	MAE	RMSE	MAE	RMSE	MAE	RMSE	MAE
LOCABAL	0.968	0.729	0.965	0.728	1.063	0.773	1.059	0.771
SREPS	0.997	0.735	0.995	0.743	1.074	0.786	1.066	0.788
HERec	0.954	0.726	0.949	0.724	1.051	0.769	1.044	0.763
SREE	0.919	0.683	0.911	0.680	1.020	0.770	1.008	0.766
CUNE	0.923	0.725	0.912	0.715	1.020	0.761	1.006	0.753
ASRNS	0.860	0.677	0.857	0.678	0.923	0.699	0.921	0.698

Tab. 7 Comparison of experimental results between ASRNS and GNN-type algorithms on all and cold start datasets

算法	全体数据集				冷启动数据集
	FilmTrust		CiaoDVD		FilmTrust		CiaoDVD
	RMSE	MAE	RMSE	MAE	RMSE	MAE	RMSE	MAE
GraphRec	0.825	0.629	0.992	0.729	0.919	0.721	1.008	0.756
ConsisRec	0.807	0.614	0.959	0.692	0.891	0.709	0.971	0.749
ASRNS（k=10）	0.786	0.605	0.906	0.675	0.857	0.678	0.921	0.698

Fig. 7 Proportion of users with different ratings

Fig. 8 Comparison of experimental results with different ratings

Fig. 9 Proportion of users with different trusts

Fig. 10 Comparison of experimental results with different trusts

References 31

1	BEHERA G， NAIN N. Handling data sparsity via item metadata embedding into deep collaborative recommender system［J］. Journal of King Saud University — Computer and Information Sciences， 2022， 34（10 Pt B）： 9953-9963. 10.1016/j.jksuci.2021.12.021
2	WANG H. MovieMat： context-aware movie recommendation with matrix factorization by matrix fitting［C］// Proceedings of the 7th International Conference on Computer and Communications. Piscataway： IEEE， 2021： 1642-1645. 10.1109/iccc54389.2021.9674549
3	RANGADURAI K， LIU Y Q， MALREDDY S， et al. NxtPost： user to post recommendations in Facebook groups［C］// Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York： ACM， 2022： 3792-3800. 10.1145/3534678.3539042
4	FENG J M， XIA Z Q， FENG X Y， et al. RBPR： a hybrid model for the new user cold start problem in recommender systems［J］. Knowledge-Based Systems， 2021， 214： No.106732. 10.1016/j.knosys.2020.106732
5	LI W T， GAO M， RONG W G， et al. Social recommendation using Euclidean embedding［C］// Proceedings of the 2017 International Joint Conference on Neural Networks. Piscataway： IEEE， 2017： 589-595. 10.1109/ijcnn.2017.7965906
6	ZHANG T W， LI W P， WANG L， et al. Social recommendation algorithm based on stochastic gradient matrix decomposition in social network［J］. Journal of Ambient Intelligence and Humanized Computing， 2020， 11（2）： 601-608. 10.1007/s12652-018-1167-7
7	BIN S， SUN G X. Matrix factorization recommendation algorithm based on multiple social relationships［J］. Mathematical Problems in Engineering， 2021， 2021： No.6610645. 10.1155/2021/6610645
8	JAMALI M， ESTER M. A matrix factorization technique with trust propagation for recommendation in social networks［C］// Proceedings of the 4th ACM Conference on Recommender Systems. New York： ACM， 2010： 135-142. 10.1145/1864708.1864736
9	吴宾，娄铮铮，叶阳东. 联合正则化的矩阵分解推荐算法［J］. 软件学报， 2018， 29（9）：2681-2696.
	WU B， LOU Z Z， YE Y D. Co-regularized matrix factorization recommendation algorithm［J］. Journal of Software， 2018， 29（9）： 2681-2696.
10	潘一腾，何发智，于海平. 一种基于信任关系隐含相似度的社会化推荐算法［J］. 计算机学报， 2018， 41（1）：65-81. 10.11897/SP.J.1016.2018.00065
	PAN Y T， HE F Z， YU H P. Social recommendation algorithm using implicit similarity in trust［J］. Chinese Journal of Computers， 2018， 41（1）： 65-81. 10.11897/SP.J.1016.2018.00065
11	TANG J L， HU X， GAO H J， et al. Exploiting local and global social context for recommendation［C］// Proceedings of the 23rd International Joint Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2013： 2712-2718.
12	GUO G B， ZHANG J， YORKE-SMITH N. TrustSVD： collaborative filtering with both the explicit and implicit influence of user trust and of item ratings［C］// Proceedings of the 29th AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2015： 123-129. 10.1609/aaai.v29i1.9153
13	XIONG F， SHEN W H， CHEN H S， et al. Exploiting implicit influence from information propagation for social recommendation［J］. IEEE Transactions on Cybernetics， 2020， 50（10）： 4186-4199. 10.1109/tcyb.2019.2939390
14	MA H， KING L， LYU M R. Learning to recommend with social trust ensemble［C］// Proceedings of the 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2009： 203-210. 10.1145/1571941.1571978
15	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. 10.11772/j.issn.1001-9081.2019030583
	TIAN B J， YANG H Y， FANG J D. Recommendation algorithm based on probability matrix factorization and fusing trust［J］. Journal of Computer Applications， 2019， 39（10）： 2834-2840. 10.11772/j.issn.1001-9081.2019030583
16	LIU C Y， ZHOU C， WU J， et al. Social recommendation with an essential preference space［C］// Proceedings of the 32nd AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2018： 346-353. 10.1609/aaai.v32i1.11245
17	CHEN L Y， ZHANG H L， WU J. Integrating dual user network embedding with matrix factorization for social recommender systems［C］// Proceedings of the 2019 International Joint Conference on Neural Networks. Piscataway： IEEE， 2019： 1-8. 10.1109/ijcnn.2019.8851715
18	WANG Y Z， SUN H B， ZHAO Y， et al. A heterogeneous graph embedding framework for location-based social network analysis in smart cities［J］. IEEE Transactions on Industrial Informatics， 2020， 16（4）： 2747-2755.
19	YU J L， YIN H Z， LI J D， et al. Self-supervised multi-channel hypergraph convolutional network for social recommendation［C］// Proceedings of the 2021 World Wide Web Conference. Republic and Canton of Geneva： International World Wide Web Conferences Steering Committee， 2021： 413-424. 10.1145/3442381.3449844
20	YU J L， YIN H Z， GAO M， et al. Socially-aware self-supervised tri-training for recommendation［C］// Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York： ACM， 2021： 2084-2092. 10.1145/3447548.3467340
21	LIAO J， ZHOU W， LUO F J， et al. SocialLGN： light graph convolution network for social recommendation［J］. Information Sciences， 2022， 589： 595-607. 10.1016/j.ins.2022.01.001
22	FAN W Q， MA Y， LI Q， et al. Graph neural networks for social recommendation［C］// Proceedings of the 2019 World Wide Web Conference. Republic and Canton of Geneva： International World Wide Web Conferences Steering Committee， 2019： 417-426. 10.1145/3308558.3313488
23	YANG L W， LIU Z W， DOU Y T， et al. ConsisRec： enhancing GNN for social recommendation via consistent neighbor aggregation［C］// Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2021： 2141-2145. 10.1145/3404835.3463028
24	PEROZZI B， AL-RFOU R， SKIENA S. DeepWalk： online learning of social representations［C］// Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2014： 701-710. 10.1145/2623330.2623732
25	GROVER A， LESKOVEC J. node2vec： Scalable feature learning for networks［C］// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2016： 855-864. 10.1145/2939672.2939754
26	ZHANG C X， YU L， WANG Y， et al. Collaborative user network embedding for social recommender systems［C］// Proceedings of the 2017 SIAM International Conference on Data Mining. Philadelphia， PA： SIAM， 2017： 381-389. 10.1137/1.9781611974973.43
27	JIANG S， LIANG B， WANG L， et al. TrustEmbed： a framework to predict ratings with trust network［C］// Proceedings of the 37th Chinese Control Conference. Piscataway： IEEE， 2018： 9672-9678. 10.23919/chicc.2018.8482820
28	XU C T， HAN K， GUI F， et al. SimilarMF： a social recommender system using an embedding method［C］// Proceedings of the 5th IEEE International Conference on High Performance Computing and Communications/ 17th IEEE International Conference on Smart City/ 5th IEEE International Conference on Data Science and Systems. Piscataway： IEEE， 2019： 1328-1334. 10.1109/hpcc/smartcity/dss.2019.00185
29	SHI C， HU B B， ZHAO W X， et al. Heterogeneous information network embedding for recommendation［J］. IEEE Transactions on Knowledge and Data Engineering， 2019， 31（2）： 357-370. 10.1109/tkde.2018.2833443
30	张青博，王斌，崔宁宁，等. 基于注意力机制的规范化矩阵分解推荐算法［J］. 软件学报， 2020， 31（3）：778-793. 10.13328/j.cnki.jos.005913
	ZHANG Q B， WANG B， CUI N N， et al. Attention-based regularized matrix factorization for recommendation［J］. Journal of Software， 2020， 31（3）： 778-793. 10.13328/j.cnki.jos.005913
31	陈嘉颖，于炯，杨兴耀，等. 考虑用户间消极相似性的排序推荐算法［J］. 计算机工程与设计， 2017， 38（5）：1247-1251， 1272. 10.16208/j.issn1000-7024.2017.05.024
	CHEN J Y， YU J， YANG X Y， et al. Ranking-based recommendation algorithm considering negative similarity between users［J］. Computer Engineering and Design， 2017， 38（5）： 1247-1251， 1272. 10.16208/j.issn1000-7024.2017.05.024

[1]	Yonggui WANG, Qiwen SHI. Social recommendation by enhanced GNN with heterogeneous relationship [J]. Journal of Computer Applications, 2023, 43(11): 3464-3471.
[2]	Hangyuan DU, Sicong HAO, Wenjian WANG. Semi-supervised representation learning method combining graph auto-encoder and clustering [J]. Journal of Computer Applications, 2022, 42(9): 2643-2651.
[3]	Le ZHOU, Tingting DAI, Chun LI, Jun XIE, Boce CHU, Feng LI, Junyi ZHANG, Qiao LIU. Network representation learning model based on node attribute bipartite graph [J]. Journal of Computer Applications, 2022, 42(8): 2311-2318.
[4]	Jun HU, Zhengkang XU, Li LIU, Fujin ZHONG. Network embedding method based on multi-granularity community information [J]. Journal of Computer Applications, 2022, 42(3): 663-670.
[5]	Weijian DENG, Xi CHEN. Containerized network embedding algorithm based on time-varying resources [J]. Journal of Computer Applications, 2022, 42(2): 550-556.
[6]	Yinying ZHOU, Mengyi ZHANG, Dunhui YU, Ming ZHU. Social recommendation combining trust implicit similarity and score similarity [J]. Journal of Computer Applications, 2022, 42(12): 3671-3678.
[7]	Qi GAO, Na LYU, Jingcheng MIAO. Wireless virtual network embedding algorithm based on load balance [J]. Journal of Computer Applications, 2022, 42(10): 3148-3153.
[8]	LI Yafang, LIANG Ye, FENG Weiwei, ZU Baokai, KANG Yujian. Deep network embedding method based on community optimization [J]. Journal of Computer Applications, 2021, 41(7): 1956-1963.
[9]	FAN Wei, WANG Huimin, XING Yan. Auto-encoder based multi-view attributed network representation learning model [J]. Journal of Computer Applications, 2021, 41(4): 1064-1070.
[10]	REN Kezhou, PENG Furong, GUO Xin, WANG Zhe, ZHANG Xiaojing. Social recommendation based on dynamic integration of social information [J]. Journal of Computer Applications, 2021, 41(10): 2806-2812.
[11]	LI Xiangkun, JIA Caiyan. Collaborative filtering method fusing overlapping community regularization and implicit feedback [J]. Journal of Computer Applications, 2021, 41(1): 53-59.
[12]	ZHAO Jihong, WU Doudou, QU Hua, YIN Zhenyu. Survivable virtual network embedding guarantee mechanism based on software defined network [J]. Journal of Computer Applications, 2020, 40(3): 770-776.
[13]	SUN Heli, HE Liang, HE Fang, SUN Miaomiao, JIA Xiaolin. Network embedding based tenuous subgraph finding [J]. Journal of Computer Applications, 2020, 40(10): 2929-2935.
[14]	HE Haochen, ZHANG Danhong. Recommendation method based on multidimensional social relationship embedded deep graph neural network [J]. Journal of Computer Applications, 2020, 40(10): 2795-2803.
[15]	YANG Yanlin, YE Zhonglin, ZHAO Haixing, MENG Lei. Link prediction algorithm based on high-order proximity approximation [J]. Journal of Computer Applications, 2019, 39(8): 2366-2373.

Adaptive social recommendation based on negative similarity

基于消极相似性的自适应社会化推荐

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 17

References 31

Related Articles 15

Recommended Articles

Metrics