TransET: Knowledge Graph Embedding with Entity Types
Abstract
:1. Introduction
2. Related Work
3. TransET Model
3.1. Preliminaries
3.2. TransET Model
4. Experiments
4.1. Datasets
- FB15K: FB15K is a widely-used dataset in knowledge graph embedding, which contains 14,951 entities, 1345 relations, and 592,213 triples. Following Bordes et al. [10], we split the triples into the training set, validation set, and test set. The entity types are collected and processed by Xie et al. [13] through , -, and - domains in Freebase;
- DBpedia98K: DBpedia98K consisting of 98,022 entities, 294 relations, and 596,797 triples is a novel dataset we built by the following steps: (1) Randomly select some relational triples from DBpedia; (2) collect the entity types through the and domains of relations contained in the triples selected; and (3) split the selected triples into thetraining set, validation set, and test set.
4.2. Link Prediction
4.2.1. Design of Experiments
4.2.2. Analysis of results
- The semantics of assistant embeddings: The assistant embeddings TransH represent the normal vectors of the relation-specific hyperplane, while in TransET represent the embeddings of entity types. The latter clarifies the semantics of the assistant embeddings, and is more targeted than the former. Thus it can better capture more precise semantic connections between entities and relations;
- The forms of mapping functions: The mapping function TransH is designed in the form of vector multiplication, while in TransET is designed in the form of circle convolution. Compared to vector multiplication, circle convolution can more directly capture the semantics between different dimensions of different vectors. For example, as shown in Figure 3, the first dimension of is the sum of the products of the same dimension of and , i.e., .
4.3. Triple Classification
Design of Experiments
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Xin, D.; Evgeniy, G.; Geremy, H.; Wilko, H.; Ni, L.; Kevin, M.; Thomas, S.; Shaohua, S.; Wei, Z. Knowledge vault: A web-scale approach to probabilistic knowledge fusion. In Proceedings of the 20th ACM SIGKDD International Conference, New York, NY, USA, 24 August 2014; pp. 601–610. [Google Scholar]
- Aidan, H.; Eva, B.; Michael, C.; Claudia, D.; Gerard, D.; Claudio, G.; Labra, G.J.E.; Sabrina, K.; Sebastian, N.; Axel, P.; et al. Knowledge graphs. arXiv 2020, arXiv:2003.02320. [Google Scholar]
- Wang, Z.; Zhang, J.; Feng, J.; Chen, Z. Knowledge graph embedding by translating on hyperplanes. In Proceedings of the 28th AAAI Conference on Artificial Intelligence, Quebec City, QC, Canada, 27–31 July 2014; pp. 1112–1119. [Google Scholar]
- Lin, Y.; Liu, Z.; Sun, M.; Liu, Y.; Zhu, X. Learning entity and relation embeddings for knowledge graph completion. In Proceedings of the 29th AAAI Conference on Artificial Intelligence, Austin, TX, USA, 25–30 January 2015; pp. 2181–2187. [Google Scholar]
- Ji, G.; Liu, K.; He, S.; Zhao, J. Knowledge graph completion with adaptive sparse transfer matrix. In Proceedings of the 30th AAAI Conference on Artificial Intelligence, Phoenix, AZ USA, 12–17 February 2016. [Google Scholar]
- Bollacker, K.; Evans, C.; Paritosh, P.; Sturge, T.; Taylor, J. Freebase: A collaboratively created graph database for structuring human knowledge. In Proceedings of the 27th ACM SIGMOD International Conference, Vancouver, BC, Canada, 9–12 June 2008. [Google Scholar]
- Auer, S.; Bizer, C.; Kobilarov, G.; Lehmann, J.; Cyganiak, R.; Ives, Z. Dbpedia: A nucleus for a web of open data. In Proceedings of the 6th International Semantic Web Conference, Busan, Korea, 11–15 November 2007. [Google Scholar]
- Krompaß, D.; Baier, S.; Tresp, V. Type-constrained representation learning in knowledge gaphs. In Proceedings of the 14th International Semantic Web Conference, Bethlehem, PA, USA, 11–15 October 2015; pp. 640–655. [Google Scholar]
- Nickel, M.; Tresp, V.; Kriegel, H. A three-way model for collective learning on multi-relational data. In Proceedings of the 28th International Conference on Machine Learning, Bellevue, WA, USA, 28 June–2 July 2011; pp. 809–816. [Google Scholar]
- Bordes, A.; Usunier, N.; Garcia-Duran, A.; Weston, J.; Yakhnenko, O. Translating embeddings for modeling multi-relational data. In Proceedings of the 27th Neural Information Processing Systems, Lake Tahoe, NV, USA, 5–10 December 2013; pp. 2787–2795. [Google Scholar]
- Guo, S.; Wang, Q.; Wang, B.; Wang, L.; Guo, L. Semantically smooth knowledge graph embedding. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics, Beijing, China, 26–31 July 2015; pp. 84–94. [Google Scholar]
- Belkin, M.; Niyogi, P. Laplacian eigenmaps and spectral techniques for embedding and clustering. In Proceedings of the 15th Neural Information Processing Systems, Vancouver, BC, Canada, 9–14 December 2001; pp. 585–591. [Google Scholar]
- Zhang, Z.; Zhuang, F.; Qu, M. Knowledge Graph embedding with hierarchical relation structure. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, 31 October–4 November 2018; pp. 3198–3207. [Google Scholar]
- Ji, G.; He, S.; Xu, L.; Liu, K.; Zhao, J. Knowledge graph embedding via dynamic mapping matrix. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics, Beijing, China, 26–31 July 2015; Volume 1, pp. 687–696. [Google Scholar]
- Song, H.; Kim, A.; Park, S. Learning translation-based knowledge graph embeddings by N-pair translation loss. J. Appl. Sci. 2020, 10, 3964. [Google Scholar] [CrossRef]
- Jenatton, R.; Le Roux, N.; Bordes, A.; Obozinski, G. A latent factor model for highly multi-relational data. In Proceedings of the 26th Annual Conference on Neural Information Processing Systems, Lake Tahoe, NV, USA, 3–6 December 2012; pp. 3176–3184. [Google Scholar]
- Yang, B.; Yih, W.T.; He, X.; Gao, J.; Deng, L. Embedding entities and relations for learning and inference in knowledge bases. In Proceedings of the 3rd International Conference on Learning Representations, San Diego, CA, USA, 7–9 May 2015. [Google Scholar]
- Nickel, M.; Rosasco, L.; Poggio, T.A. Holographic embeddings of knowledge graphs. In Proceedings of the 30th AAAI Conference on Artificial Intelligence, Phoenix, AZ, USA, 12–17 February 2016; pp. 1955–1961. [Google Scholar]
- Lin, Y.; Liu, Z.; Sun, M. Knowledge representation learning with entities, attributes and relations. In Proceedings of the 25th International Joint Conference on Artificial Intelligence, New York, NY, USA, 9–15 July 2016; pp. 2866–2872. [Google Scholar]
- Xie, R.; Liu, Z.; Sun, M. Representation learning of knowledge graphs with hierarchical types. In Proceedings of the 25th International Joint Conference on Artificial Intelligence, New York, NY, USA, 9–15 July 2016; pp. 2965–2971. [Google Scholar]
- Tang, X.; Chen, L.; Cui, J. Knowledge representation learning with entity descriptions, hierarchical types, and textual relations. J. Inf. Process. Manag. 2019, 56, 809–822. [Google Scholar] [CrossRef]
- Zhou, J.; Wang, P.; Pan, Z.; Xu, Z. JECI: A joint knowledge graph embedding model for concepts and instances. In Proceedings of the 9th JIST Conference, Hangzhou, China, 25–27 November 2019; pp. 82–98. [Google Scholar]
- Zhou, J.; Wang, P. JECI++: A modified joint knowledge graph embedding model for concepts and instances. J. Big Data Res. 2020, 24, 100160. [Google Scholar]
- Rahman, M.; Takasu, A. Knowledge graph embedding via entities’ type mapping matrix. In Proceedings of the 25th ICONIP, Siem Reap, Cambodia, 13–16 December 2018. [Google Scholar]
- Balkır, E.; Naslidnyk, M.; Palfrey, D.; Mittal, A. Improving knowledge graph embeddings with inferred entity types. In Proceedings of the 32nd Annual Conference on Neural Information Processing Systems, Montreal, QC, Canada, 3–8 December 2018. [Google Scholar]
- Kim, K.; Hur, Y.; Kim, G.; Lim, H. GREG: A global level relation extraction with knowledge graph embedding. J. Appl. Sci. 2020, 10, 1181. [Google Scholar] [CrossRef] [Green Version]
- Su, C.; Song, H.; Park, S. An approach to knowledge base completion by a committee-based knowledge graph embedding. J. Appl. Sci. 2020, 10, 2651. [Google Scholar]
- Bordes, A.; Weston, J.; Collobert, R.; Bengio, Y. Learning structured embeddings of knowledge bases. In Proceedings of the 25th AAAI Conference on Artificial Intelligence, San Francisco, CA, USA, 7–8 August 2011; p. 6. [Google Scholar]
- Van der Maaten, L.; Hinton, G. Visualizing data using t-SNE. J. Mach. Res. 2008, 9, 2579–2605. [Google Scholar]
Dataset | FB15K | DBpedia98K |
---|---|---|
#Entity | 14,951 | 98,022 |
#Relation | 1345 | 294 |
#Entity Type | 571 | 91 |
#Training Set | 483,142 | 530,797 |
#Validation Set | 50,000 | 60,000 |
#Test Set | 59,071 | 60,000 |
Model | Space Complexity | Time Complexity |
---|---|---|
TransE | ||
TransH | ||
TransR | ||
TransD | ||
KR-EAR | ||
TKRL | ||
TransET |
Dataset | FB15K | DBpedia98K | ||||||
---|---|---|---|---|---|---|---|---|
Metric Setting | MR | Hits@10(%) | MR | Hits@10(%) | ||||
Raw | Filter | Raw | Filter | Raw | Filter | Raw | Filter | |
TransE | 243 | 125 | 34.9 | 47.1 | 262 | 148 | 30.2 | 45.1 |
TransH | 211 | 84 | 42.5 | 58.5 | 258 | 119 | 35.6 | 53.4 |
TransR | 226 | 78 | 43.8 | 65.5 | 235 | 83 | 38.8 | 55.6 |
TransD | 211 | 67 | 49.4 | 74.2 | 217 | 65 | 44.1 | 60.4 |
KR-EAR | 118 | 39.5 | 57.3 | 213 | 71 | 33.5 | 48.9 | |
TKRL | 202 | 87 | 73.4 | 209 | 63 | 46.4 | 63.4 | |
TransET | 187 | 50.1 |
Task | Predicting Head Entity | Predicting Tail Entity | ||||||
---|---|---|---|---|---|---|---|---|
Type | 1-to-1 | 1-to-N | N-to-1 | N-to-N | 1-to-1 | 1-to-N | N-to-1 | N-to-N |
TransE | 43.7 | 65.7 | 18.2 | 47.2 | 43.7 | 19.7 | 66.7 | 50.0 |
TransH | 66.7 | 81.7 | 30.2 | 57.4 | 63.7 | 30.1 | 83.2 | 60.8 |
TransR | 76.9 | 77.9 | 38.1 | 66.9 | 76.2 | 38.4 | 76.2 | 69.1 |
TransD | 80.7 | 85.8 | 47.1 | 75.6 | 80.0 | 54.5 | 80.7 | 77.9 |
TransET |
Dataset | FB15K | DBpedia98K |
---|---|---|
TransE | 77.6 | 70.3 |
TransH | 77.1 | 69.2 |
TransR | 79.8 | 74.5 |
TransD | 84.2 | 78.8 |
KR-EAR | 85.7 | 77.4 |
TKRL | 88.5 | 80.2 |
TransET |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Wang, P.; Zhou, J.; Liu, Y.; Zhou, X. TransET: Knowledge Graph Embedding with Entity Types. Electronics 2021, 10, 1407. https://doi.org/10.3390/electronics10121407
Wang P, Zhou J, Liu Y, Zhou X. TransET: Knowledge Graph Embedding with Entity Types. Electronics. 2021; 10(12):1407. https://doi.org/10.3390/electronics10121407
Chicago/Turabian StyleWang, Peng, Jing Zhou, Yuzhang Liu, and Xingchen Zhou. 2021. "TransET: Knowledge Graph Embedding with Entity Types" Electronics 10, no. 12: 1407. https://doi.org/10.3390/electronics10121407