基于多侧面信息表征联合的实体相似性度量及对齐方法

doi:10.19678/j.issn.1000-3428.0068839

摘要/Abstract

摘要：

实体对齐旨在发现不同知识图谱中相同对象的不同实例, 但图谱之间的异构性导致等价实例结构及表征不一致, 从而影响实体对齐准确性。提出一种实体主信息与多侧面信息表征相联合的异构图谱实体相似性度量方法, 并用于实体对齐任务。实体主信息包括实体名称及描述, 侧面信息包括实体属性、关系及关联实体描述等信息。针对图谱间等价实体结构异构带来的对齐干扰, 提出了一种结合实体多侧面信息语义表征的相似性度量方法UnMuSIR-SM&EA用于实体对齐。为提升信息同义词的表示一致性, 引入表示学习模型以获取实体各信息的语义表征, 为解决表示学习模型嵌入空间各向异性带来的同义词度量尺度不一致问题, 设计了一种基于实体主信息对比学习的微调方法, 优化实体信息的语义表征。实验结果表明, 该方法在结构差异较大的数据集DIS_ZH-EN上的Hits@1达到了95.2%, 比基于侧面信息的模型BERT-INT高出了16.8百分点; 在DBP15K的DBP15K_ZH-EN、DBP15K_JA-EN和DBP15K_FR-EN数据子集上的Hits@1分别达到了95.7%、96.0%和98.9%;在DBP-WD数据集上的Hits@1达到了99.4%。所提模型在实体对齐任务上具有优异的效果。

关键词: 实体对齐, 知识图谱, 相似性度量, 对比学习, 预训练模型

Abstract:

Entity alignment aims to identify different instances of the same object in different knowledge graphs. However, the heterogeneity between graphs leads to inconsistent equivalent instance structures and representations, thereby affecting the accuracy of entity alignment. A heterogeneous graph entity similarity measurement method that combines the main entity information and multi-side information representation is proposed and applied to entity alignment tasks. The main entity information includes the entity name and description, while the peripheral information includes entity attributes, relationships, and related entity descriptions. A similarity metric method called UnMuSIR-SM&EA, which combines the semantic representations of the multi-side information of entities, is proposed for entity alignment to address the alignment interference caused by the heterogeneity of equivalent entity structures between graphs. A representation learning model is introduced to obtain the semantic representations of various types of entity information, to improve synonym representation consistency. To solve the problem of inconsistent synonym measurement scales caused by spatial anisotropy in the embeddings of representation learning models, a fine-tuning method is designed based on entity main information contrastive learning to optimize the semantic representation of entity information. Experimental results show that the proposed method performs exceptionally well on the DIS_ZH-EN dataset with significant structural differences with Hits@1 reaching 95.2%, which is 16.8 percentage points higher than the BERT-INT model based on peripheral information. On the DBP15K_ZH-EN, DBP15K_JA-EN, and DBP15K_FR-EN subsets of DBP15K data Hits@1 is 95.7%, 96.0%, and 98.9%, respectively. On the DBP-WD dataset Hits@1 is 99.4%. The proposed model exhibites excellent performance in entity alignment tasks.

Key words: entity alignment, knowledge graphs, similarity metrics, comparative learning, pre-training model

朱红, 王阔然, 朱彤. 基于多侧面信息表征联合的实体相似性度量及对齐方法[J]. 计算机工程, 2025, 51(3): 64-75.

ZHU Hong, WANG Kuoran, ZHU Tong. Entity Similarity Metrics and Alignment Method Based on the Union of Multi-Side Information Representations[J]. Computer Engineering, 2025, 51(3): 64-75.

https://www.ecice06.com/CN/Y2025/V51/I3/64

图/表 10

图1 UnMuSIR-SM&EA整体框架

Fig.1 Overall framework of UnMuSIR-SM&EA

图2 实体信息表征学习模块结构

Fig.2 Structure of entity information representation learning module

图3 侧面信息相似度特征聚合计算流程

Fig.3 Aggregation calculation procedure of side information similarity feature

图4 模型训练损失曲线

Fig.4 Loss curves of model training

参考文献 28

1	ZHANG X F , ZHANG R C , CHEN J F , et al. Semi-supervised entity alignment with global alignment and local information aggregation. IEEE Transactions on Knowledge and Data Engineering, 2023, 35 (10): 10464- 10477. doi: 10.1109/TKDE.2023.3238993
2	ZHU B B , BAO T , WANG K R , et al. A semi-supervised neighborhood matching model for global entity alignment. Neural Computing and Applications, 2023, 35 (15): 10779- 10799. doi: 10.1007/s00521-023-08264-y
3	DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional Transformers for language understanding[EB/OL]. [2023-10-11]. https://arxiv.org/pdf/1810.04805.
4	黄峻福, 李天瑞, 贾真, 等. 中文异构百科知识库实体对齐. 计算机应用, 2016, 36 (7): 1881-1886, 1898. doi: 10.11772/j.issn.1001-9081.2016.07.1881
	HUANG J F , LI T R , JIA Z , et al. Enity alignment of Chinese heterogeneous encyclopedia knowledge base. Journal of Computer Applications, 2016, 36 (7): 1881-1886, 1898. doi: 10.11772/j.issn.1001-9081.2016.07.1881
5	朱蓓蓓. 面向知识图谱的实体对齐研究[D]. 长春: 吉林大学, 2023.
	ZHU B B. Research on entity alignment for knowledge graphs[D]. Changchun: Jilin University, 2023. (in Chinese)
6	CHEN M H, TIAN Y T, YANG M H, et al. Multi-aspect information knowledge graph embeddings for cross-lingual knowledge alignment[C]//Proceedings of Conference on Empirical Methods in Natural Language Processing. Melbourne, Australia: AAAI Press, 2017: 1511-1517.
7	ZHU H, XIE R B, LIU Z Y. Iterative entity alignment via joint knowledge embeddings[C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence. New York, USA: ACM Press, 2017: 4258-4264.
8	CAO Y X, LIU Z Y, LI C J, et al. Multi-channel graph neural network for entity alignment[EB/OL]. [2023-10-11]. https://arxiv.org/abs/1908.09898.
9	刘雪丽, 李燕, 李春雨, 等. 基于多方嵌入的逐步实体对齐方法. 现代电子技术, 2024, 47 (13): 138- 143. doi: 10.16652/j.issn.1004-373x.2024.13.025
	LIU X L , LI Y , LI C Y , et al. Stepwise entity alignment based on multi-party embedding. Modern Electronics Technique, 2024, 47 (13): 138- 143. doi: 10.16652/j.issn.1004-373x.2024.13.025
10	WANG Z C, LV Q S, LAN X H, et al. Cross-lingual knowledge graph alignment via graph convolutional networks[C]//Proceedings of Conference on Empirical Methods in Natural Language Processing. Brussels, Belgium: Association for Computational Linguistics, 2018: 349-357.
11	SUN Z W, HU W, LI C K, et al. Cross-lingual entity alignment via joint attribute-preserving embedding[EB/OL]. [2023-10-11]. https://arxiv.org/pdf/1708.05045.
12	WU Y T, LIU X, FENG Y S, et al. Neighborhood matching network for entity alignment[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Brussels, Belgium: Association for Computational Linguistics, 2020: 6477-6487.
13	CHEN M H, TIAN Y T, CHANG, K W, et al. Co-training embeddings of knowledge graphs and entity descriptions for cross-lingual entity alignment[EB/OL]. [2023-10-11]. https://arxiv.org/pdf/1806.06478.
14	WU Y T, LIU X, FENG Y S, et al. Relation-aware entity alignment for heterogeneous knowledge graphs[EB/OL]. [2023-10-11]. https://arxiv.org/abs/1908.08210.
15	YANG H W, ZOU Y Y, SHI P, et al. Aligning cross-lingual entities with multi-aspect information[EB/OL]. [2023-10-11]. https://arxiv.org/abs/1910.06575.
16	LAI K H, ZHA D, LI Y N, et al. BERT-INT: a BERT-based interaction model for knowledge graph alignment[C]//Proceedings of International Joint Conference on Artificial Intelligence. Montreal, Canada: AAAI Press, 2021: 3174-3180.
17	李文娜, 张智雄. 基于联合语义表示的不同知识库中的实体对齐方法研究. 数据分析与知识发现, 2021, 5 (7): 1- 9. doi: 10.11925/infotech.2096-3467.2021.0143
	LI W N , ZHANG Z X . Entity alignment method for different knowledge repositories with joint semantic representation. Data Analysis and Knowledge Discovery, 2021, 5 (7): 1- 9. doi: 10.11925/infotech.2096-3467.2021.0143
18	ETHAYARAJH K. How contextual are contextualized word representations? comparing the geometry of BERT, ELMo, and GPT-2 embeddings[EB/OL]. [2023-10-11]. https://arxiv.org/abs/1909.00512v1.
19	PETERS M E, NEUMANN M, IYYER M, et al. Deep contextualized word representations[EB/OL]. [2023-10-11]. https://arxiv.org/pdf/1802.05365.
20	RADFORD A, WU J, CHILD R, et al. Language models are unsupervised multitask learners[EB/OL]. [2023-10-11]. https://cdn.openai.com/better-language-models/language_models_are_unsupervised_multitask_learners.pdf.
21	WANG T Z, ISOLA P. Understanding contrastive representation learning through alignment and uniformity on the hypersphere[C]//Proceedings of International Conference on Machine Learning. [S. l. ]: AAAI Press, 2021: 9309-10077.
22	LI B H, ZHOU H, HE J X, et al. On the sentence embeddings from pre-trained language models[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing. New York, USA: Association for Computational Linguistics, 2020: 9119-9130.
23	SU J L, CAO J R, LIU W J, et al. Whitening sentence representations for better semantics and faster retrieval[EB/OL]. [2023-10-11]. https://arxiv.org/abs/2103.15316?context=cs.CL.
24	GAO T Y, YAO X C, CHEN D Q, et al. SimCSE: simple contrastive learning of sentence embeddings[EB/OL]. [2023-10-11]. https://arxiv.org/abs/2104.08821?context=cs.LG.
25	GUO M H , LIU Z N , MU T J , et al. Beyond self-attention: external attention using two linear layers for visual tasks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45 (5): 5436- 5447.
26	HE K M, FAN H Q, WU Y X, et al. Momentum contrast for unsupervised visual representation learning[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Washington D.C., USA: IEEE Press, 2019: 9726-9735.
27	XIONG C Y, DAI Z Y, CALLAN J, et al. End-to-end neural ad-hoc ranking with kernel pooling[C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM Press, 2017: 55-64.
28	SUN Z Q, HU W, ZHANG Q H, et al. Bootstrapping entity alignment with knowledge graph embedding[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence. New York, USA: ACM Press, 2018: 4396-4402.

[1]	刘春雨, 陈庆锋, 莫少聪, 谢泽. 基于逻辑规则和图神经网络的知识图谱补全[J]. 计算机工程, 2025, 51(3): 131-143.
[2]	郭俊辰, 马御棠, 相艳, 赵学东, 郭军军. 基于Prompt打分的实体链接方法[J]. 计算机工程, 2025, 51(3): 334-341.
[3]	王庆丰, 李旭, 姚春龙, 程腾腾. 面向研究生招生咨询的中文Text-to-SQL模型[J]. 计算机工程, 2025, 51(3): 362-368.
[4]	马恒志, 钱育蓉, 冷洪勇, 吴海鹏, 陶文彬, 张依杨. 知识图谱嵌入研究进展综述[J]. 计算机工程, 2025, 51(2): 18-34.
[5]	饶东宁, 许正辉, 梁瑞仕. 基于知识库问答的回答生成研究[J]. 计算机工程, 2025, 51(2): 94-101.
[6]	姚利峰, 蔡满春, 朱懿, 陈咏豪, 张溢文. 基于字节编码与预训练任务的加密流量分类模型[J]. 计算机工程, 2025, 51(2): 188-201.
[7]	费涛, 艾山·吾买尔, 杜文旭, 朱翠翠. 基于Squeezeformer的多颗粒度多方面发音质量评测方法[J]. 计算机工程, 2025, 51(1): 81-87.
[8]	魏嵬, 丁香香, 郭梦星, 杨钊, 刘辉. 文本相似度计算方法综述[J]. 计算机工程, 2024, 50(9): 18-32.
[9]	李泽霖, 吕兆峰, 陈富强, 李克. 基于多跳信息融合的实体对齐模型[J]. 计算机工程, 2024, 50(9): 142-152.
[10]	汤志康, 武毓琦, 李春英, 汤庸. 基于知识图谱卷积网络的学习资源推荐[J]. 计算机工程, 2024, 50(9): 153-160.
[11]	李华昱, 张智康, 闫阳, 岳阳. 基于知识图谱增强的领域多模态实体识别[J]. 计算机工程, 2024, 50(8): 31-39.
[12]	高爽, 史轶伦, 徐巧枝, 于磊. 基于对比学习的非对称编解码结构的心脏MRI分割研究[J]. 计算机工程, 2024, 50(8): 290-300.
[13]	李惠乾, 钟柏昌. 教育知识图谱: 研究进展与未来发展——基于2013—2023年中文核心期刊载文的分析[J]. 计算机工程, 2024, 50(7): 1-12.
[14]	王炼红, 林飞鹏, 李潇瑶, 谌桂枝, 周莉. 融入课程知识图谱的KMAKT预测[J]. 计算机工程, 2024, 50(7): 23-31.
[15]	钱来, 赵卫伟. 基于对比学习和注意力机制的文本分类方法[J]. 计算机工程, 2024, 50(7): 104-111.

选择文件类型/文献管理软件名称

选择包含的内容