Abstract
Entity linking plays a vital role in Question Answering over Knowledge Graphs (KGQA), and the representation of entities is a fundamental component of entity linking for user questions. In order to alleviate the problem of entity descriptions that unrelated texts obfuscate similar entities, we present a new entity linking framework, which refines the encodings of entity descriptions based on entity difference modeling, so that entity linking’s ability to distinguish among similar entities is improved. The entity differences are modeled in a two-stage approach: the initial differences are first computed among similar entity candidates by comparing their descriptions, and then interactions between the initial differences and questions are performed to extract key differences, which identify critical information for entity linking. On the basis of the key differences, subsequent entity description encodings are refined, and entity linking is then performed using the refined entity representations. Experimental results on end-to-end benchmark datasets demonstrate that our approach achieves state-of-the-art precision, recall and F1-score.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Model available at https://huggingface.co/bert-large-uncased/tree/main.
- 2.
Code, data and model available at https://github.com/facebookresearch/BLINK/tree/master/elq.
References
Banerjee, D., Chaudhuri, D., Dubey, M., Lehmann, J.: PNEL: pointer network based end-to-end entity linking over knowledge graphs. In: Pan, J.Z., Tamma, V., d’Amato, C., Janowicz, K., Fu, B., Polleres, A., Seneviratne, O., Kagal, L. (eds.) ISWC 2020. LNCS, vol. 12506, pp. 21–38. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-62419-4_2
Bordes, A., Usunier, N., Garcia-Durán, A., Weston, J., Yakhnenko, O.: Translating embeddings for modeling multi-relational data. In: 26th International Conference on Neural Information Processing Systems, pp. 2787–2795 (2013)
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. In: 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 4171–4186 (2019)
Ferragina, P., Scaiella, U.: Fast and accurate annotation of short texts with wikipedia pages. IEEE Softw. 29(1), 70–75 (2011)
Gu, Y., et al.: Beyond IID: three levels of generalization for question answering on knowledge bases. In: Web Conference 2021, pp. 3477–3488 (2021)
Johnson, J., Douze, M., Jégou, H.: Billion-scale similarity search with GPUs. IEEE Transactions on Big Data (2019)
Li, B.Z., Min, S., Iyer, S., Mehdad, Y., Yih, W.T.: Efficient one-pass end-to-end entity linking for questions. In: 2020 Conference on Empirical Methods in Natural Language Processing, pp. 6433–6441 (2020)
Sakor, A., et al.: Old is gold: linguistic driven approach for entity and relation linking of short text. In: 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 2336–2346 (2019)
Sakor, A., Singh, K., Patel, A., Vidal, M.E.: Falcon 2.0: an entity and relation linking tool over wikidata. In: 29th ACM International Conference on Information & Knowledge Management, pp. 3141–3148 (2020)
Singh, K., Lytra, I., Radhakrishna, A.S., Shekarpour, S., Vidal, M.E., Lehmann, J.: No one is perfect: analysing the performance of question answering components over the dbpedia knowledge graph. J. Web Semant. 65, 100594 (2020)
Singh, K., et al.: Why reinvent the wheel: let’s build question answering systems together. In: 2018 World Wide Web Conference, pp. 1247–1256 (2018)
Sorokin, D., Gurevych, I.: Mixing context granularities for improved entity linking on question answering data across entity categories. In: 7th Joint Conference on Lexical and Computational Semantics, pp. 65–75 (2018)
Su, Y., et al.: On generating characteristic-rich question sets for QA evaluation. In: 2016 Conference on Empirical Methods in Natural Language Processing, pp. 562–572 (2016)
Vinyals, O., Fortunato, M., Jaitly, N.: Pointer networks. Adv. Neural Info. Proc. Syst. 28, 2692–2700 (2015)
Wu, L., Petroni, F., Josifoski, M., Riedel, S., Zettlemoyer, L.: Scalable zero-shot entity linking with dense entity retrieval. In: 2020 Conference on Empirical Methods in Natural Language Processing, pp. 6397–6407 (2020)
Yih, W.T., Richardson, M., Meek, C., Chang, M.W., Suh, J.: The value of semantic parse labeling for knowledge base question answering. In: 54th Annual Meeting of the Association for Computational Linguistics, pp. 201–206 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wang, M., Li, M., Sun, K., Hou, Z. (2022). Entity Difference Modeling Based Entity Linking for Question Answering over Knowledge Graphs. In: Lu, W., Huang, S., Hong, Y., Zhou, X. (eds) Natural Language Processing and Chinese Computing. NLPCC 2022. Lecture Notes in Computer Science(), vol 13551. Springer, Cham. https://doi.org/10.1007/978-3-031-17120-8_18
Download citation
DOI: https://doi.org/10.1007/978-3-031-17120-8_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-17119-2
Online ISBN: 978-3-031-17120-8
eBook Packages: Computer ScienceComputer Science (R0)