《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (11): 3464-3471.DOI: 10.11772/j.issn.1001-9081.2022111774
• 数据科学与技术 • 上一篇
收稿日期:
2022-11-28
修回日期:
2023-03-22
接受日期:
2023-03-23
发布日期:
2023-04-07
出版日期:
2023-11-10
通讯作者:
时启文
作者简介:
王永贵(1967—),男,内蒙古赤峰人,教授,硕士,CCF会员,主要研究方向:大数据、并行计算、数据库、数据挖掘Received:
2022-11-28
Revised:
2023-03-22
Accepted:
2023-03-23
Online:
2023-04-07
Published:
2023-11-10
Contact:
Qiwen SHI
About author:
WANG Yonggui, born in 1967, M. S., professor. His research interests include big data, parallel computing, database, data mining.Supported by:
摘要:
社交推荐旨在利用用户的社会属性推荐潜在的感兴趣项目,有效缓解了数据稀疏性和冷启动问题。然而现有的社交推荐算法主要面向单一社交关系进行研究,社会属性难以充分参与计算,存在未能合理利用社会异构关系和节点特征表示质量不高的问题,为此提出一种结合异构关系增强图神经网络的社交推荐模型(HR-GNN)。HR?GNN利用图卷积网络(GCN)聚合用户和项目节点信息,生成查询嵌入以查询节点信息;通过将抽样概率与邻居节点之间的一致性分数相结合的邻居抽样策略挖掘社会异构关系;用自注意力机制聚合节点信息以提高用户和项目特征表示的质量。在两个真实数据集上进行的实验结果表明,所提算法在平均绝对误差(MAE)和均方根误差(RMSE)两个指标上相较于基准算法均有明显改进,在Ciao数据集上它们分别至少降低了1.80%和1.35%,在Epinions数据集上则分别至少降低了2.80%和3.18%,验证了HR-GNN的有效性。
中图分类号:
王永贵, 时启文. 结合异构关系增强图神经网络的社交推荐[J]. 计算机应用, 2023, 43(11): 3464-3471.
Yonggui WANG, Qiwen SHI. Social recommendation by enhanced GNN with heterogeneous relationship[J]. Journal of Computer Applications, 2023, 43(11): 3464-3471.
数据集 | 用户数 | 项目数 | 用户-项目评分数 | 用户-用户社交数 |
---|---|---|---|---|
Ciao | 7 374 | 105 059 | 282 163 | 111 781 |
Epinions | 26 337 | 139 738 | 664 832 | 487 182 |
表1 数据集统计结果
Tab. 1 Statistics of datasets
数据集 | 用户数 | 项目数 | 用户-项目评分数 | 用户-用户社交数 |
---|---|---|---|---|
Ciao | 7 374 | 105 059 | 282 163 | 111 781 |
Epinions | 26 337 | 139 738 | 664 832 | 487 182 |
序号 | Ciao | Epinions | ||
---|---|---|---|---|
MAE | RMSE | MAE | RMSE | |
均值 | 0.719 4 | 0.949 5 | 0.777 2 | 1.015 0 |
1 | 0.719 7 | 0.949 8 | 0.771 2 | 1.014 8 |
2 | 0.720 1 | 0.949 5 | 0.782 6 | 1.015 4 |
3 | 0.719 3 | 0.949 3 | 0.771 7 | 1.014 4 |
4 | 0.718 7 | 0.949 1 | 0.775 7 | 1.015 2 |
5 | 0.719 4 | 0.949 7 | 0.785 1 | 1.015 6 |
表2 HR-GNN模型运行5次的均值结果
Tab. 2 Mean results of HR-GNN model running 5 times
序号 | Ciao | Epinions | ||
---|---|---|---|---|
MAE | RMSE | MAE | RMSE | |
均值 | 0.719 4 | 0.949 5 | 0.777 2 | 1.015 0 |
1 | 0.719 7 | 0.949 8 | 0.771 2 | 1.014 8 |
2 | 0.720 1 | 0.949 5 | 0.782 6 | 1.015 4 |
3 | 0.719 3 | 0.949 3 | 0.771 7 | 1.014 4 |
4 | 0.718 7 | 0.949 1 | 0.775 7 | 1.015 2 |
5 | 0.719 4 | 0.949 7 | 0.785 1 | 1.015 6 |
算法 | Ciao | Epinions | ||
---|---|---|---|---|
MAE | RMSE | MAE | RMSE | |
PMF | 0.902 1 | 1.123 8 | 0.995 2 | 1.212 8 |
SocialMF | 0.832 1 | 1.065 7 | 0.883 7 | 1.132 8 |
SoReg | 0.898 7 | 1.094 7 | 0.941 2 | 1.193 6 |
SAMN | 0.811 6 | 1.092 4 | 0.899 5 | 1.189 9 |
EATNN | 0.797 3 | 1.074 2 | 0.866 3 | 1.138 5 |
GraphRec | 0.759 1 | 1.009 3 | 0.844 1 | 1.087 8 |
Danser | 0.740 3 | 0.990 4 | 0.813 9 | 1.075 6 |
ASR | ||||
ConsisRec | 0.739 4 | 0.972 2 | 0.804 6 | 1.049 5 |
HR-GNN | 0.719 4 | 0.949 5 | 0.777 2 | 1.015 0 |
表3 两个数据集上不同模型的实验结果
Tab. 3 Experimental results of different models on two datasets
算法 | Ciao | Epinions | ||
---|---|---|---|---|
MAE | RMSE | MAE | RMSE | |
PMF | 0.902 1 | 1.123 8 | 0.995 2 | 1.212 8 |
SocialMF | 0.832 1 | 1.065 7 | 0.883 7 | 1.132 8 |
SoReg | 0.898 7 | 1.094 7 | 0.941 2 | 1.193 6 |
SAMN | 0.811 6 | 1.092 4 | 0.899 5 | 1.189 9 |
EATNN | 0.797 3 | 1.074 2 | 0.866 3 | 1.138 5 |
GraphRec | 0.759 1 | 1.009 3 | 0.844 1 | 1.087 8 |
Danser | 0.740 3 | 0.990 4 | 0.813 9 | 1.075 6 |
ASR | ||||
ConsisRec | 0.739 4 | 0.972 2 | 0.804 6 | 1.049 5 |
HR-GNN | 0.719 4 | 0.949 5 | 0.777 2 | 1.015 0 |
模型 | 数据集 | 平均训练时间/s |
---|---|---|
GraphRec | Ciao | 221.84 |
Epinions | 587.04 | |
Danser | Ciao | 315.46 |
Epinions | 663.12 | |
ASR | Ciao | 269.31 |
Epinions | 729.63 | |
ConsisRec | Ciao | 1 500.55 |
Epinions | 2 818.32 | |
HR-GNN | Ciao | 256.81 |
Epinions | 621.24 |
表4 不同模型在两个数据集上的平均训练时间
Tab. 4 Average training time of different models on two datasets
模型 | 数据集 | 平均训练时间/s |
---|---|---|
GraphRec | Ciao | 221.84 |
Epinions | 587.04 | |
Danser | Ciao | 315.46 |
Epinions | 663.12 | |
ASR | Ciao | 269.31 |
Epinions | 729.63 | |
ConsisRec | Ciao | 1 500.55 |
Epinions | 2 818.32 | |
HR-GNN | Ciao | 256.81 |
Epinions | 621.24 |
模型 | 参数量/106 | 模型大小/MB | 占用内存/GB |
---|---|---|---|
GraphRec | 3.88 | 14.81 | 1.49 |
Danser | 3.85 | 14.68 | 1.42 |
ASR | 3.91 | 14.91 | 1.56 |
ConsisRec | 3.82 | 14.56 | 1.38 |
HR-GNN | 3.86 | 14.72 | 1.44 |
表5 不同模型的参数量分析
Tab. 5 Parametric quantitative analysis of different models
模型 | 参数量/106 | 模型大小/MB | 占用内存/GB |
---|---|---|---|
GraphRec | 3.88 | 14.81 | 1.49 |
Danser | 3.85 | 14.68 | 1.42 |
ASR | 3.91 | 14.91 | 1.56 |
ConsisRec | 3.82 | 14.56 | 1.38 |
HR-GNN | 3.86 | 14.72 | 1.44 |
1 | YANG B, LEI Y, LIU J, et al. Social collaborative filtering by trust[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(8): 1633-1647. 10.1109/tpami.2016.2605085 |
2 | 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 |
3 | ABADAL S, JAIN A, GUIRADO R, et al. Computing graph neural networks: a survey from algorithms to accelerators[J]. ACM Computing Surveys, 2022, 54(9): No.191. 10.1145/3477141 |
4 | FAN W, LI Q, CHENG M. Deep modeling of social relations for recommendation[C]// Proceedings of the 32nd AAAI Conference on Artificial Intelligence. Palo Alto, CA: AAAI Press, 2018: 8075-8076. 10.1609/aaai.v32i1.12132 |
5 | LIU Z, DOU Y, YU P S, et al. Alleviating the inconsistency problem of applying graph neural network to fraud detection[C]// Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 1569-1572. 10.1145/3397271.3401253 |
6 | LIU Z, LI X, FAN Z, et al. Basket recommendation with multi-intent translation graph neural network[C]// Proceedings of the 2020 IEEE International Conference on Big Data. Piscataway: IEEE, 2020: 728-737. 10.1109/bigdata50022.2020.9377917 |
7 | KRASNOVA H, WIDJAJA T, BUXMANN P, et al. Research note — why following friends can hurt you: an exploratory investigation of the effects of envy on social networking sites among college-age users[J]. Information Systems Research, 2015, 26(3): 585-605. 10.1287/isre.2015.0588 |
8 | SONG L, BI Y, YAO M, et al. DREAM: a dynamic relation-aware model for social recommendation[C]// Proceedings of the 29th ACM International Conference on Information and Knowledge Management. New York: ACM, 2020: 2225-2228. 10.1145/3340531.3412115 |
9 | FU M, QU H, YI Z, et al. A novel deep learning-based collaborative filtering model for recommendation system[J]. IEEE Transactions on Cybernetics, 2019, 49(3): 1084-1096.. 10.1109/tcyb.2018.2795041 |
10 | NI J, HUANG Z, CHENG J, et al. An effective recommendation model based on deep representation learning[J]. Information Sciences, 2021, 542: 324-342. 10.1016/j.ins.2020.07.038 |
11 | BAHDANAU D, CHO K, BENGIO Y. Neural machine translation by jointly learning to align and translate[EB/OL]. (2016-05-19) [2022-09-13].. 10.1017/9781108608480.003 |
12 | WU L, SUN P, FU Y, et al. A neural influence diffusion model for social recommendation[C]// Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2019: 235-244. 10.1145/3331184.3331214 |
13 | CHEN C, ZHANG M, LIU Y, et al. Social attentional memory network: modeling aspect-and friend-level differences in recommendation[C]// Proceedings of the 12th ACM International Conference on Web Search and Data Mining. New York: ACM, 2019: 177-185. 10.1145/3289600.3290982 |
14 | CHEN C, ZHANG M, WANG C, et al. An efficient adaptive transfer neural network for social-aware recommendation[C]// Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2019: 225-234. 10.1145/3331184.3331192 |
15 | GAO C, WANG X, HE X, et al. Graph neural networks for recommender system[C]// Proceedings of the 15th ACM International Conference on Web Search and Data Mining. New York: ACM, 2022: 1623-1625. 10.1145/3488560.3501396 |
16 | WU S, SUN F, ZHANG W, et al. Graph neural networks in recommender systems: a survey[J]. ACM Computing Surveys, 2023, 55(5): No.97. 10.1145/3535101 |
17 | 黄立威,江碧涛,吕守业,等. 基于深度学习的推荐系统研究综述[J]. 计算机学报, 2018, 41(7):1619-1647. 10.11897/SP.J.1016.2018.01619 |
HUANG L W, JIANG B T, LYU S Y, et al. Survey on deep learning based recommender systems[J]. Chinese Journal of Computers, 2018, 41(7): 1619-1647. 10.11897/SP.J.1016.2018.01619 | |
18 | FAN W, 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 |
19 | WU L, LI J, SUN P, et al. DiffNet++: a neural influence and interest diffusion network for social recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(10): 4753-4766. 10.1109/tkde.2020.3048414 |
20 | WU Q, ZHANG H, GAO X, et al. Dual graph attention networks for deep latent representation of multifaceted social effects in recommender systems[C]// Proceedings of the 2019 World Wide Web Conference. Republic and Canton of Geneva: International World Wide Web Conferences Steering Committee, 2019: 2091-2102. 10.1145/3308558.3313442 |
21 | LUO D, BIAN Y, ZHANG X, et al. Attentive social recommendation: Towards user and item diversities[EB/OL]. [2022-09-13].. 10.48550/arXiv.2011.04797 |
22 | XU H, HUANG C, XU Y, et al. Global context enhanced social recommendation with hierarchical graph neural networks[C]// Proceedings of the 2020 IEEE International Conference on Data Mining. Piscataway: IEEE, 2020: 701-710. 10.1109/icdm50108.2020.00079 |
23 | YANG L, LIU Z, DOU Y, 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 | CONG W, FORSATI R, KANDEMIR M, et al. Minimal variance sampling with provable guarantees for fast training of graph neural networks[C]// Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2020: 1393-1403. 10.1145/3394486.3403192 |
25 | ZENG H, ZHOU H, SRIVASTAVA A, et al. GraphSAINT: graph sampling based inductive learning method[EB/OL]. (2020-02-16) [2022-09-13].. |
26 | CHIANG W L, LIU X, SI S, et al. Cluster-GCN: an efficient algorithm for training deep and large graph convolutional networks[C]// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 257-266. 10.1145/3292500.3330925 |
27 | SALAKHUTDINOV R, MNIH A. Probabilistic matrix factorization[C]// Proceedings of the 20th International Conference on Neural Information Processing Systems. Red Hook, NY: Curran Associates Inc., 2007: 1257-1264. 10.1145/1390156.1390267 |
28 | MA H, ZHOU D, LIU C, et al. Recommender systems with social regularization[C]// Proceedings of the 4th ACM International Conference on Web Search and Data Mining. New York: ACM, 2011: 287-296. 10.1145/1935826.1935877 |
[1] | 王宏, 钱清, 王欢, 龙永. 融合大核注意力卷积的轻量化图像篡改定位算法[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2692-2699. |
[2] | 李众, 王雅婧, 马巧梅. 基于空洞卷积的医学图像超分辨率重建算法[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2940-2947. |
[3] | 张秋余, 温永旺. 用于语音检索的三联体深度哈希方法[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2910-2918. |
[4] | 徐丽, 符祥远, 李浩然. 基于门控卷积的时空交通流预测模型[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2760-2765. |
[5] | 杨昊, 张轶. 基于上下文信息和多尺度融合重要性感知的特征金字塔网络算法[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2727-2734. |
[6] | 袁国龙, 张玉金, 刘洋. 基于残差反馈和自注意力的图像篡改取证网络[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2925-2931. |
[7] | 段升位, 程欣宇, 王浩舟, 王飞. 基于改进的YOLOv5的大坝表面病害检测算法[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2619-2629. |
[8] | 崔雨萌, 王靖亚, 刘晓文, 闫尚义, 陶知众. 融合注意力和裁剪机制的通用文本分类模型[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2396-2405. |
[9] | 齐爱玲, 王宣淋. 基于中层细微特征提取与多尺度特征融合细粒度图像识别[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2556-2563. |
[10] | 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. |
[11] | 金泽熙, 李磊, 刘继. 基于改进领域分离网络的迁移学习模型[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2382-2389. |
[12] | 王静红, 周志霞, 王辉, 李昊康. 双路自编码器的属性网络表示学习[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2338-2344. |
[13] | 刘源, 董永权, 贾瑞, 杨昊霖. 面向个性化课程推荐的分层分期注意力网络模型[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2358-2363. |
[14] | 梁美佳, 刘昕武, 胡晓鹏. 基于改进YOLOv3的列车运行环境图像小目标检测算法[J]. 《计算机应用》唯一官方网站, 2023, 43(8): 2611-2618. |
[15] | 魏远, 林彦, 郭晟楠, 林友芳, 万怀宇. 融合出发地与目的地时空相关性的城市区域间出租车需求预测[J]. 《计算机应用》唯一官方网站, 2023, 43(7): 2100-2106. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||