Entity Alignment Method Based on Neighborhood Aggregation

doi:10.19678/j.issn.1000-3428.0061589

Abstract

Abstract: Entity Alignment(EA) aims to judge whether entities from different Knowledge Graph(KG) are the same object pointing to the real world.However, the structural heterogeneity between KG often affects the accuracy of EA.Hence, an EA method based on a Neighborhood Aggregation Matching Network(NAMN) model is proposed.Based on the different importance of each hop neighbor to the central entity, a hierarchical idea is applied to process the neighborhood information of each hop differently, and the gating mechanism is used to perform aggregation to learn the representation of a graph structure.Subsequently, a neighborhood local subgraph is constructed for each entity for cross graph neighborhood matching, and the output of the matching stage is jointly encoded with the graph structure representation learned through the gating mechanism to generate the final matching oriented representation.The experiment is performed using the DBP15K dataset.The experimental results show that all values of Hits@1 exceed 75%, all values of Hits@10 are between 85% and 97%, and the Mean Reciprocal Rank(MMR) exceeds 80%, indicating that the NAMN model can effectively improve the matching accuracy of entities.

Key words: Entity Alignment(EA), Knowledge Graph(KG), gated neighborhood aggregation, neighborhood matching, alignment prediction

摘要： 实体对齐旨在判断来自不同知识图谱的实体是否为指向真实世界的同一个对象。然而，知识图谱间的结构异质性往往会影响实体对齐的准确性。提出一种基于邻域聚合匹配网络（NAMN）模型的实体对齐方法。根据每跳邻居对中心实体重要性不同的特点，采用分层的思想区别处理每跳邻域信息，通过门控机制进行聚合以学习图结构的表征。在此基础上，为每个实体构建邻域局部子图进行跨图邻域匹配，并将匹配阶段的输出与通过门控机制所学习到的图结构表征进行联合编码，生成最终面向匹配的表征。采用DBP15K数据集进行实验，结果显示，Hits@1的所有值均在75%以上，Hits@10的所有值均在85%以上，最高可达到97%，平均倒数排名均高于80%，表明NAMN模型能够有效提高实体的匹配准确度。

关键词: 实体对齐, 知识图谱, 门控邻域聚合, 邻域匹配, 对齐预测

CLC Number:

TP391

TAN Yuanzhen, LI Xiaonan, LI Guanyu. Entity Alignment Method Based on Neighborhood Aggregation[J]. Computer Engineering, 2022, 48(6): 65-72.

谭元珍, 李晓楠, 李冠宇. 基于邻域聚合的实体对齐方法[J]. 计算机工程, 2022, 48(6): 65-72.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0061589

http://www.ecice06.com/EN/Y2022/V48/I6/65

Figures/Tables 11

References

[1] 孟明明, 张坤, 论兵, 等.一种面向知识图谱问答的语义查询扩展方法[J].计算机工程, 2019, 45(9):276-283, 290. MENG M M, ZHANG K, LUN B, et al.A semantic query expansion method for question answering based on knowledge graph[J].Computer Engineering, 2019, 45(9):276-283, 290.(in Chinese)
[2] 王辉, 郁波, 洪宇, 等.基于知识图谱的Web信息抽取系统[J].计算机工程, 2017, 43(6):118-124. WANG H, YU B, HONG Y, et al.Web information extraction system based on knowledge graph[J].Computer Engineering, 2017, 43(6):118-124.(in Chinese)
[3] CAO Y X, HOU L, LI J Z, et al.Joint representation learning of cross-lingual words and entities via attentive distant supervision[C]//Proceedings of 2018 Conference on Empirical Methods in Natural Language Processing.Stroudsburg, USA:Association for Computational Linguistics, 2018:1-11.
[4] CAO Y X, WANG X, HE X N, et al.Unifying knowledge graph learning and recommendation:towards a better understanding of user preferences[C]//Proceedings of World Wide Web Conference.Washington D.C., USA:IEEE Press, 2019:151-161.
[5] BIZER C, LEHMANN J, KOBILAROV G, et al.DBpedia-a crystallization point for the Web of data[J].Journal of Web Semantics, 2009, 7(3):154-165.
[6] SUCHANEK F M, KASNECI G, WEIKUM G.YAGO:a large ontology from Wikipedia and WordNet[J].Journal of Web Semantics, 2008, 6(3):203-217.
[7] SPEER R, CHIN J, HAVASI C.ConceptNet 5.5:an open multilingual graph of general knowledge[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence.[S.l.]:AAAI, 2017:4444-4451.
[8] CARLSON A, BETTERIDGE J, KISIEL B, et al.Toward an arcHitsecture for never-ending language learning[C]//Proceedings of AAAI Conference on Artificial Intelligence.[S.l.]:AAAI, 2010:1-10.
[9] MAHDISOLTANI F, BIEGA J, SUCHANEK F.YAGO3:a knowledge base from multilingual Wikipedias[C]//Proceedings of the 7th Biennial Conference on Innovative Data Systems Research.Asilomar, USA:[s.n.], 2014:4-7.
[10] CHEN M H, TIAN Y T, YANG M H, et al.Multi-lingual knowledge graph embeddings for cross-lingual knowledge alignment[EB/OL].[2021-05-05]. https://arxiv.org/abs/1611.03954.
[11] GLOROT X, BENGIO Y.Understanding the difficulty of training deep feedforward neural networks[C]//Proceedings of the 13th International Conference on Artificial Intelligence and Statistics.[S.l.]:JMLR, 2010:249-256.
[12] SUN Z, HU W, ZHANG Q, et al.Bootstrapping entity alignment with knowledge graph embedding[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Washington D.C., USA:IEEE Press, 2018:4396-4402.
[13] SUN Z, HU W, LI C.Cross-lingual entity alignment via joint attribute-preserving embedding[C]//Proceedings of International Semantic Web Conference.Berlin, Germany:Springer, 2017:628-644.
[14] GUO L, SUN Z, HU W.Learning to exploit long-term relational dependencies in knowledge graphs[C]//Proceedings of International Conference on Machine Learning.[S.l.]:PMLR, 2019:2505-2514.
[15] ZHANG Q H, SUN Z Q, HU W, et al.Multi-view knowledge graph embedding for entity alignment[EB/OL].[2021-05-05]. https://arxiv.org/abs/1906.02390.
[16] WANG Z C, LV Q S, LAN X H, et al.Cross-lingual knowledge graph alignment via graph convolutional networks[C]//Proceedings of 2018 Conference on Empirical Methods in Natural Language Processing.Stroudsburg, USA:Association for Computational Linguistics, 2018:349-357.
[17] SCHLICHTKRULL M, KIPF T N, BLOEM P, et al.Modeling relational data with graph convolutional networks[C]//Proceedings of European Semantic Web Conference.Berlin, Germany:Springer, 2018:593-607.
[18] WU Y, LIU X, FENG Y, et al.Relation-aware entity alignment for heterogeneous knowledge graphs[EB/OL].[2021-04-10].https://arxiv.org/abs/1908.08210.
[19] YING R, YOU J, MORRIS C, et al.Hierarchical graph representation learning with differentiable pooling[EB/OL].[2021-05-05].https://proceedings.neurips.cc/paper/2018/hash/e77dbaf6759253c7c6d0efc5690369c7-Abstract.html.
[20] MORRIS C, RITZERT M, FEY M, et al.Weisfeiler and leman go neural:higher-order graph neural networks[C]//Proceedings of 2019 AAAI Conference on Artificial Intelligence.[S.l.]:AAAI, 2019:4602-4609.
[21] LEMAN A A, WEISFEILER B.A reduction of a graph to a canonical form and an algebra arising during this reduction[J].Nauchno-Technicheskaya Informatsiya, 1968, 2(9):12-16.
[22] PUJARA J, MIAO H, GETOOR L, et al.Knowledge graph identification[C]//Proceedings of International Semantic Web Conference.Berlin, Germany:Springer, 2013:542-557.
[23] SUN Z, WANG C, HU W, et al.Knowledge graph alignment network with gated multi-hop neighborhood aggregation[C]//Proceedings of 2020 AAAI Conference on Artificial Intelligence.[S.l.]:AAAI, 2020:222-229.
[24] KIPF T N, WELLING M.Semi-supervised classification with graph convolutional networks[EB/OL].[2021-05-05].https://arxiv.53yu.com/abs/1609.02907.
[25] LI Y, GU C, DULLIEN T, et al.Graph matching networks for learning the similarity of graph structured objects[C]//Proceedings of International Conference on Machine Learning.[S.l.]:PMLR, 2019:3835-3845.
[26] XU K, WANG L, YU M, et al.Cross-lingual knowledge graph alignment via graph matching neural network[EB/OL].[2021-05-05].https://arxiv.53yu.com/abs/1905. 11605.
[27] SRIVASTAVA R K, GREFF K, SCHMIDHUBER J.Highway networks[EB/OL].[2021-05-05].https://arxiv.53yu.com/abs/1505.00387.
[28] VELIČKOVIĆ P, CUCURULL G, CASANOVA A, et al.Graph attention networks[EB/OL].[2021-05-05].https://arxiv.53yu.com/abs/1710.10903.
[29] WU Y, LIU X, FENG Y, et al.Neighborhood matching network for entity alignment[EB/OL].[2021-05-05].https://arxiv.53yu.com/abs/2005.05607.
[30] LI Y J, TARLOW D, BROCKSCHMIDT M, et al.Gated graph sequence neural networks[EB/OL].[2021-05-05]. https://arxiv.org/abs/1511.05493.
[31] KINGMA D P, BA J.Adam:a method for stochastic optimization[EB/OL].[2021-05-05].https://arxiv.53yu.com/abs/1412.6980.
[32] CAO Y, LIU Z, LI C, et al.Multi-channel graph neural network for entity alignment[EB/OL].[2021-05-05]. https://arxiv.org/abs/1905.11605.
[33] LEHMANN J, ISELE R, JAKOB M, et al.DBpedia-a large-scale, multilingual knowledge base extracted from Wikipedia[J].Semantic Web, 2015, 6(2):167-195.

Please choose a citation manager

Content to export