Recommendation method based on multi-perspective relation-enhanced knowledge graph

doi:10.11772/j.issn.1001-9081.2024111665

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (11): 3519-3528.DOI: 10.11772/j.issn.1001-9081.2024111665

• Artificial intelligence • Previous Articles

Recommendation method based on multi-perspective relation-enhanced knowledge graph

Ke GAN, Xiaofei ZHU(), Jiawei CHENG

College of Computer Science and Engineering，Chongqing University of Technology，Chongqing 401320，China

Received:2024-11-27 Revised:2025-04-11 Accepted:2025-04-18 Online:2025-04-22 Published:2025-11-10
Contact: Xiaofei ZHU
About author:GAN Ke， born in 2000， M. S. candidate. His research interests include natural language processing， knowledge graph recommendation.
CHENG Jiawei， born in 1999， M. S. candidate. His research interests include natural language processing， emotion recognition.
Supported by:
National Natural Science Foundation of China(62472059);Natural Science Foundation of Chongqing(CSTB2022NSCQ-LZX0002);Chongqing Talent Plan Program(CSTC2024YCJH-BGZXM0022);Major Project of Science and Technology Research Plan of Chongqing Municipal Education Commission(KJZD-M202201102)

基于多视角关系增强知识图谱的推荐方法

甘轲, 朱小飞(), 程佳玮

重庆理工大学计算机科学与工程学院，重庆 401320

通讯作者: 朱小飞
作者简介:甘轲（2000—），男，四川广安人，硕士研究生，主要研究方向：自然语言处理、知识图谱推荐
程佳玮（1999—），男，江西上饶人，硕士研究生，主要研究方向：自然语言处理、情感识别。
基金资助:
国家自然科学基金资助项目(62472059);重庆市自然科学基金资助项目(CSTB2022NSCQ-LZX0002);重庆市自然科学基金资助项目(CSTB2022NSCQ-MSX1672);重庆英才计划项目(CSTC2024YCJH-BGZXM0022);重庆市教育委员会科学技术研究计划重大项目(KJZD-M202201102)

Abstract

Abstract:

Knowledge graph-based recommendation methods learn representations of user and item nodes by integrating relational connections from both item-attribute graph and user-interaction graph to recommend suitable items. However， since knowledge graphs contain both noisy and high-quality relations， the primary challenge lies in circumventing noisy relations while effectively mining valuable relations. Existing methods typically rely on global reconstruction strategies to optimize knowledge graph relations through single manner of pruning noisy relations or mining high-quality relations， thereby learning user and item representations. However， global perspectives struggle to sufficiently capture detailed local information and tend to overlook potential complementarity between local and global information. Furthermore， solely employing pruning or supplementation strategies fail to simultaneously mitigate interference of noisy relations and comprehensively exploit high-quality relations. To address these issues， a Recommendation method based on Multi-Perspective Relation-Enhanced Knowledge Graph （RMPREKG） was proposed. This method alleviated noise impacts in both item-attribute graphs and user?interaction graphs through an item hybrid relation alignment module and an interaction hybrid relation enhancement module， while deeply exploring valuable high-order relations. The item hybrid relation alignment module extracted local and global relations separately via importance pruning strategy and high-order relation mining approach， followed by a knowledge alignment method to synergize both types of information for effectively refining high-quality item auxiliary information. The interaction hybrid relation enhancement module constructed local hybrid pruning relation graph and global hybrid supplementation relation graph， with cross-channel and cross-layer contrastive learning employed to enhance their informational complementarity， thereby comprehensively learning user and item representations. Finally， a hierarchical gated adaptive fusion method was adopted to integrate multiple sets of user and item embeddings for recommendation. When the recommendation length is 20， compared to VRKG4Rec （Virtual Relational Knowledge Graphs for Recommendation）， RMPREKG achieves a 10.17% improvement in Normalized Discounted Cumulative Gain （NDCG） on the Last.FM dataset and a 1.13% improvement in NDCG on the MovieLens-1M dataset.

Key words: knowledge graph, multi-perspective, knowledge alignment, contrastive learning, gate fusion

摘要：

基于知识图谱的推荐方法通过结合物品属性图和用户交互图中的关系连接学习用户和物品节点的表示，从而推荐合适的物品；然而，由于知识图谱同时包含噪声关系和优质关系，当前的主要挑战在于如何规避噪声关系并挖掘优质关系。现有方法通常基于全局的重构策略，通过裁剪噪声关系或挖掘优质关系的单一方式优化知识图谱关系，以此学习用户和物品的表示。然而，基于全局视角难以充分捕捉局部信息的细节，并且容易忽略局部信息与全局信息之间的潜在互补性；此外，仅依赖裁剪或增补策略难以同时规避噪声关系的干扰并全面挖掘优质关系。针对上述问题，提出一种基于多视角关系增强知识图谱的推荐方法（RMPREKG）。该方法利用物品混合关系对齐模块和交互混合关系增强模块减少物品属性图和用户交互图中噪声关系的影响，同时深入挖掘优质高阶关系。物品混合关系对齐模块通过重要性裁剪策略和高阶关系挖掘方法分别提取局部与全局关系，并采用一种知识对齐的方法协同两类信息，可有效提炼优质物品辅助信息；交互混合关系增强模块构建了局部混合裁剪关系图和全局混合增补关系图，并通过跨通道和跨层对比学习增强两者之间的信息互补性，从而全面学习用户和物品的表示。最后，采用层级门控自适应的方法融合多组用户与物品嵌入用于推荐。当推荐长度为20时，与VRKG4Rec（Virtual Relational Knowledge Graphs for Recommendation）相比，RMPREKG在Last.FM数据集的归一化折损累计增益（NDCG）提升了10.17%，在MovieLens-1M数据集的NDCG提升了1.13%。

关键词: 知识图谱, 多视角, 知识对齐, 对比学习, 门控融合

CLC Number:

TP391.3

Ke GAN, Xiaofei ZHU, Jiawei CHENG. Recommendation method based on multi-perspective relation-enhanced knowledge graph[J]. Journal of Computer Applications, 2025, 45(11): 3519-3528.

甘轲, 朱小飞, 程佳玮. 基于多视角关系增强知识图谱的推荐方法[J]. 《计算机应用》唯一官方网站, 2025, 45(11): 3519-3528.

Figures/Tables 11

References 30

[1]	赵俊逸，庄福振，敖翔，等.协同过滤推荐系统综述［J］.信息安全学报，2021，6（5）：17-34.
	ZHAO J Y， ZHUANG F Z， AO X， et al. Survey of collaborative filtering recommender systems［J］. Journal of Cyber Security， 2021， 6（5）： 17-34.
[2]	GUO Q， ZHUANG F， QIN C， et al. A survey on knowledge graph-based recommender systems［J］. IEEE Transactions on Knowledge and Data Engineering， 2022， 34（8）： 3549-3568.
[3]	WANG H， ZHAO M， XIE X， et al. Knowledge graph convolutional networks for recommender systems［C］// Proceedings of the 2019 World Wide Web Conference. New York： ACM， 2019： 3307-3313.
[4]	JIANG Y， YANG Y， XIA L， et al. DiffKG： knowledge graph diffusion model for recommendation［C］// Proceedings of the 17th ACM International Conference on Web Search and Data Mining. New York： ACM， 2024： 313-321.
[5]	WANG X， HE X， CAO Y， et al. KGAT： knowledge graph attention network for recommendation［C］// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2019： 950-958.
[6]	WANG S， YANG L， GONG J， et al. MetaKRec： collaborative meta-knowledge enhanced recommender system［C］// Proceedings of the 2022 IEEE International Conference on Big Data. Piscataway： IEEE， 2022： 665-674.
[7]	ZHU X， DU Y， MAO Y， et al. Knowledge-refined denoising network for robust recommendation［C］// Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2023： 362-371.
[8]	LU L， WANG B， ZHANG Z， et al. VRKG4Rec： virtual relational knowledge graph for recommendation［C］// Proceedings of the 16th ACM International Conference on Web Search and Data Mining. New York： ACM， 2023： 526-534.
[9]	XIAO H， HUANG M， ZHU X. TransG： a generative mixture model for knowledge graph embedding［C］// Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics （Volume 1： Long Papers）. Stroudsburg： ACL， 2016： 2316-2325.
[10]	BORDES A， USUNIER N， GARCIA-DURÁN A， et al. Translating embeddings for modeling multi-relational data［C］// Proceedings of the 27th International Conference on Neural Information Processing Systems — Volume 2. Red Hook： Curran Associates Inc.， 2013： 2787-2795.
[11]	WANG Z， ZHANG J， FENG J， et al. Knowledge graph embedding by translating on hyperplanes［C］// Proceedings of the 28th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2014： 1112-1119.
[12]	JI G， HE S， XU L， et al. Knowledge graph embedding via dynamic mapping matrix［C］// Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics （Volume 1： Long Papers）. Stroudsburg： ACL， 2015： 687-696.
[13]	XIAO H， HUANG M， HAO Y， et al. TransA： an adaptive approach for knowledge graph embedding［EB/OL］. ［2024-04-11］. .
[14]	WANG H， ZHANG F， WANG J， et al. RippleNet： propagating user preferences on the knowledge graph for recommender systems［C］// Proceedings of the 27th ACM International Conference on Information and Knowledge Management. New York： ACM， 2018： 417-426.
[15]	CHEN H， LI Y， SUN X， et al. Temporal meta-path guided explainable recommendation［C］// Proceedings of the 14th ACM International Conference on Web Search and Data Mining. New York： ACM， 2021： 1056-1064.
[16]	ZOU D， WEI W， WANG Z， et al. Improving knowledge-aware recommendation with multi-level interactive contrastive learning［C］// Proceedings of the 31st ACM International Conference on Information and Knowledge Management. New York： ACM， 2022： 2817-2826.
[17]	ZOU D， WEI W， MAO X L， et al. Multi-level cross-view contrastive learning for knowledge-aware recommender system［C］// Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2022： 1358-1368.
[18]	YANG Y， HUANG C， XIA L， et al. Knowledge graph contrastive learning for recommendation［C］// Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2022： 1434-1443.
[19]	YU J， YIN H， XIA X， et al. Are graph augmentations necessary？ simple graph contrastive learning for recommendation［C］// Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2022： 1294-1303.
[20]	CHEN M， HUANG C， XIA L， et al. Heterogeneous graph contrastive learning for recommendation［C］// Proceedings of the 16th ACM International Conference on Web Search and Data Mining. New York： ACM， 2023： 544-552.
[21]	TU K， CUI P， WANG D， et al. Conditional graph attention networks for distilling and refining knowledge graphs in recommendation［C］// Proceedings of the 30th ACM International Conference on Information and Knowledge Management. New York： ACM， 2021： 1834-1843.
[22]	YANG Y， HUANG C， XIA L， et al. Knowledge graph self-supervised rationalization for recommendation［C］// Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York： ACM， 2023： 3046-3056.
[23]	WU J， WANG X， FENG F， et al. Self-supervised graph learning for recommendation［C］// Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2021： 726-735.
[24]	FAN Z， XU K， DONG Z， et al. Graph collaborative signals denoising and augmentation for recommendation［C］// Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2023： 2037-2041.
[25]	CAI X， HUANG C， XIA L， et al. LightGCL： simple yet effective graph contrastive learning for recommendation［EB/OL］. ［2024-01-03］. .
[26]	HE W， SUN G， LU J， et al. Candidate-aware graph contrastive learning for recommendation［C］// Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2023： 1670-1679.
[27]	HE X， CHUA T S. Neural factorization machines for sparse predictive analytics［C］// Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2017： 355-364.
[28]	ZHANG F， YUAN N J， LIAN D， et al. Collaborative knowledge base embedding for recommender systems［C］// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2016： 353-362.
[29]	ZOU D， WEI W， ZHU F， et al. Knowledge enhanced multi-intent transformer network for recommendation［C］// Companion Proceedings of the ACM Web Conference 2024. New York： ACM， 2024： 1-9.
[30]	WANG X， HUANG T， WANG D， et al. Learning intents behind interactions with knowledge graph for recommendation［C］// Proceedings of the Web Conference 2021. New York： ACM， 2021： 878-887.

数据集	模型	Metric@1			Metric@5			Metric@10			Metric@20
数据集	模型	Recall	NDCG	HR	Recall	NDCG	HR	Recall	NDCG	HR	Recall	NDCG	HR
Last.FM	NFM	1.50	3.90	3.90	5.95	4.80	13.20	9.52	6.26	23.10	14.97	8.05	29.90
	CKE	4.43	10.31	10.31	13.06	11.26	26.58	18.85	13.62	25.02	26.95	16.25	46.11
	KGAT	2.41	5.67	5.67	7.86	9.49	16.74	12.56	12.58	25.92	20.59	16.71	37.67
	KGTN	4.71	10.90	10.90	13.40	14.30	27.37	19.59	14.12	37.14	28.06	16.87	48.87
	KGIN	6.06	13.98	13.98	17.42	15.24	35.92	24.96	18.32	47.07	35.49	21.69	59.07
	VRKG4Rec	6.79	16.34	16.34	20.15	17.62	39.84	28.05	20.85	50.69	38.78	23.02	61.84
	RMPREKG	7.23	17.30	17.30	21.12	18.56	41.47	29.73	22.08	53.04	39.69	25.36	63.89
ML	NFM	2.98	27.70	27.70	11.62	24.88	62.40	17.88	23.80	74.40	27.59	24.84	84.30
	CKE	3.85	33.54	33.54	13.62	28.75	66.65	21.19	27.89	78.29	31.30	29.18	86.51
	KGAT	2.63	23.15	23.15	10.03	20.68	57.01	16.59	21.09	71.59	26.37	23.05	82.15
	KGTN	2.97	27.55	27.55	10.63	23.76	59.90	17.40	23.80	72.84	26.91	24.76	82.67
	KGIN	4.69	11.99	11.99	15.14	12.92	31.22	22.66	15.95	43.22	31.50	19.35	53.22
	VRKG4Rec	4.29	36.74	36.74	15.01	31.38	70.13	23.29	30.53	80.55	34.12	31.92	88.34
	RMPREKG	4.32	37.28	37.28	15.18	31.81	70.54	23.63	30.89	81.89	34.46	32.28	88.39

数据集	模型	Metric@1			Metric@5			Metric@10			Metric@20
数据集	模型	Recall	NDCG	HR	Recall	NDCG	HR	Recall	NDCG	HR	Recall	NDCG	HR
Last.FM	NFM	1.50	3.90	3.90	5.95	4.80	13.20	9.52	6.26	23.10	14.97	8.05	29.90
	CKE	4.43	10.31	10.31	13.06	11.26	26.58	18.85	13.62	25.02	26.95	16.25	46.11
	KGAT	2.41	5.67	5.67	7.86	9.49	16.74	12.56	12.58	25.92	20.59	16.71	37.67
	KGTN	4.71	10.90	10.90	13.40	14.30	27.37	19.59	14.12	37.14	28.06	16.87	48.87
	KGIN	6.06	13.98	13.98	17.42	15.24	35.92	24.96	18.32	47.07	35.49	21.69	59.07
	VRKG4Rec	6.79	16.34	16.34	20.15	17.62	39.84	28.05	20.85	50.69	38.78	23.02	61.84
	RMPREKG	7.23	17.30	17.30	21.12	18.56	41.47	29.73	22.08	53.04	39.69	25.36	63.89
ML	NFM	2.98	27.70	27.70	11.62	24.88	62.40	17.88	23.80	74.40	27.59	24.84	84.30
	CKE	3.85	33.54	33.54	13.62	28.75	66.65	21.19	27.89	78.29	31.30	29.18	86.51
	KGAT	2.63	23.15	23.15	10.03	20.68	57.01	16.59	21.09	71.59	26.37	23.05	82.15
	KGTN	2.97	27.55	27.55	10.63	23.76	59.90	17.40	23.80	72.84	26.91	24.76	82.67
	KGIN	4.69	11.99	11.99	15.14	12.92	31.22	22.66	15.95	43.22	31.50	19.35	53.22
	VRKG4Rec	4.29	36.74	36.74	15.01	31.38	70.13	23.29	30.53	80.55	34.12	31.92	88.34
	RMPREKG	4.32	37.28	37.28	15.18	31.81	70.54	23.63	30.89	81.89	34.46	32.28	88.39

模型	Last.FM			ML
模型	Recall	NDCG	HR	Recall	NDCG	HR
RMPREKG	39.69	25.36	63.89	34.46	32.28	88.39
w/o align	39.09	24.76	62.68	33.74	31.65	87.63
w/o mask	39.43	24.83	63.08	34.33	32.17	88.23
w/o ecl	38.86	24.69	62.16	33.31	30.96	87.54
w/o layer	39.30	24.61	62.84	34.25	32.18	88.18
w/o channel	39.42	25.03	63.04	34.30	32.17	88.27

模型	Last.FM			ML
模型	Recall	NDCG	HR	Recall	NDCG	HR
RMPREKG	39.69	25.36	63.89	34.46	32.28	88.39
w/o align	39.09	24.76	62.68	33.74	31.65	87.63
w/o mask	39.43	24.83	63.08	34.33	32.17	88.23
w/o ecl	38.86	24.69	62.16	33.31	30.96	87.54
w/o layer	39.30	24.61	62.84	34.25	32.18	88.18
w/o channel	39.42	25.03	63.04	34.30	32.17	88.27

数据集	不同聚类簇数时的HR@20/%
数据集	1	2	3	4	5	6
Last.FM	63.38	63.46	63.64	63.89	63.74	63.53
ML	88.17	88.22	88.25	88.29	88.39	88.28

Recommendation method based on multi-perspective relation-enhanced knowledge graph

基于多视角关系增强知识图谱的推荐方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 30

Related Articles 15

Recommended Articles

Metrics

融合方式	Last.FM		ML
融合方式	Recall@20	NDCG@20	Recall@20	NDCG@20
Concate	26.97	15.73	29.96	27.02
Average	34.37	21.37	28.30	26.81
Gated Fusion	39.69	25.36	34.46	32.28

[1]	Zhixiong XU, Bo LI, Xiaoyong BIAN, Qiren HU. Adversarial sample embedded attention U-Net for 3D medical image segmentation [J]. Journal of Computer Applications, 2025, 45(9): 3011-3016.
[2]	Chao LIU, Yanhua YU. Knowledge-aware recommendation model combining denoising strategy and multi-view contrastive learning [J]. Journal of Computer Applications, 2025, 45(9): 2827-2837.
[3]	Zhiyuan WANG, Tao PENG, Jie YANG. Integrating internal and external data for out-of-distribution detection training and testing [J]. Journal of Computer Applications, 2025, 45(8): 2497-2506.
[4]	Jin XIE, Surong CHU, Yan QIANG, Juanjuan ZHAO, Hua ZHANG, Yong GAO. Dual-branch distribution consistency contrastive learning model for hard negative sample identification in chest X-rays [J]. Journal of Computer Applications, 2025, 45(7): 2369-2377.
[5]	Zhenzhou WANG, Fangfang GUO, Jingfang SU, He SU, Jianchao WANG. Robustness optimization method of visual model for intelligent inspection [J]. Journal of Computer Applications, 2025, 45(7): 2361-2368.
[6]	Chaoying JIANG, Qian LI, Ning LIU, Lei LIU, Lizhen CUI. Readmission prediction model based on graph contrastive learning [J]. Journal of Computer Applications, 2025, 45(6): 1784-1792.
[7]	Mingfeng YU, Yongbin QIN, Ruizhang HUANG, Yanping CHEN, Chuan LIN. Multi-label text classification method based on contrastive learning enhanced dual-attention mechanism [J]. Journal of Computer Applications, 2025, 45(6): 1732-1740.
[8]	Shuang LIU, Daqing LIU, Jiana MENG, Di ZHAO. Hyper-relational knowledge graph completion method fusing noise filtering [J]. Journal of Computer Applications, 2025, 45(6): 1817-1826.
[9]	Wenjing YAN, Ruidong WANG, Min ZUO, Qingchuan ZHANG. Recipe recommendation model based on hierarchical learning of flavor embedding heterogeneous graph [J]. Journal of Computer Applications, 2025, 45(6): 1869-1878.
[10]	Yufei LONG, Yuchen MOU, Ye LIU. Multi-source data representation learning model based on tensorized graph convolutional network and contrastive learning [J]. Journal of Computer Applications, 2025, 45(5): 1372-1378.
[11]	Wenbin HU, Tianxiang CAI, Tianle HAN, Zhaoman ZHONG, Changxia MA. Multimodal sarcasm detection model integrating contrastive learning with sentiment analysis [J]. Journal of Computer Applications, 2025, 45(5): 1432-1438.
[12]	Sheping ZHAI, Qing YANG, Yan HUANG, Rui YANG. Knowledge graph completion using hierarchical attention fusing directed relationships and relational paths [J]. Journal of Computer Applications, 2025, 45(4): 1148-1156.
[13]	Liqin WANG, Zhilei GENG, Yingshuang LI, Yongfeng DONG, Meng BIAN. Open-world knowledge reasoning model based on path and enhanced triplet text [J]. Journal of Computer Applications, 2025, 45(4): 1177-1183.
[14]	Cong WANG, Yancui SHI. Group recommendation model by graph neural network based on multi-perspective learning [J]. Journal of Computer Applications, 2025, 45(4): 1205-1212.
[15]	Zixin XU, Xiuwen YI, Jie BAO, Tianrui LI, Junbo ZHANG, Yu ZHENG. Construction and application of knowledge graph for epidemiological investigation [J]. Journal of Computer Applications, 2025, 45(4): 1340-1348.