Abstract
Entity linking (EL) is a fundamental task in natural language processing. Based on neural networks, existing systems pay more attention to the construction of the global model, but ignore latent semantic information in the local model and the acquisition of effective entity type information. In this paper, we propose two adaptive features, in which the first adaptive feature enables the local and global models to capture latent information, and the second adaptive feature describes effective information for entity type embeddings. These adaptive features can work together naturally to handle some uncertain entity type information for EL. Experimental results demonstrate that our EL system achieves the best performance on the AIDA-B and MSNBC datasets, and the best average performance on out-domain datasets. These results indicate that the proposed adaptive features, which are based on their own diverse contexts, can capture information that is conducive for EL.
摘要
实体链接是自然语言处理中的一项基本任务。现有的基于神经网络的系统更多地关注全局模型的构建,而忽略了局部模型中潜在的语义信息和有效实体类型信息的获取。本文提出两个自适应特征,其中第一个自适应特征使得局部和全局模型能够捕获潜在信息,第二个自适应特征能够描述实体类型嵌入的有效信息。这些自适应特征可以很自然地协同工作来处理一些不确定的实体类型信息。实验结果表明,我们的实体链接系统在AIDA-B和MSNBC数据集上取得了最佳的性能,并在域外数据集上达到了最佳的平均性能。这些结果表明,所提出的自适应特征能够基于其自身不同的上下文来捕获有利于实体链接的信息。
Similar content being viewed by others
References
Bunescu R, Paşca M, 2006. Using encyclopedic knowledge for named entity disambiguation. Proc 11th Conf of the European Chapter of the Association for Computational Linguistics, p.9–16.
Cao YX, Hou L, Li JZ, et al., 2018. Neural collective entity linking. Proc 27th Int Conf on Computational Linguistics, p.675–686.
Chen S, Wang JP, Jiang F, et al., 2020. Improving entity linking by modeling latent entity type information. Proc 34th AAAI Conf on Artificial Intelligence, p.7529–7537. https://doi.org/10.1609/aaai.v34i05.6251
Deng ZH, Li ZX, Yang Q, et al., 2020. Improving entity linking with graph networks. Proc 21st Int Conf on Web Information Systems Engineering, p.343–354. https://doi.org/10.1007/978-3-030-62005-9_25
Devlin J, Chang MW, Lee K, et al., 2019. BERT: pre-training of deep bidirectional transformers for language understanding. Proc Conf of the North American Chapter of the Association for Computational Linguistics, p.4171–4186. https://doi.org/10.18653/v1/N19-1423
Durrett G, Klein D, 2014. A joint model for entity analysis: coreference, typing, and linking. Trans Assoc Comput Linguist, 2:477–490. https://doi.org/10.1162/tacl_a_00197
Fang W, Zhang JW, Wang DL, et al., 2016. Entity disambiguation by knowledge and text jointly embedding. Proc 20th SIGNLL Conf on Computational Natural Language Learning, p.260–269. https://doi.org/10.18653/v1/K16-1026
Fang Z, Cao YN, Li Q, et al., 2019. Joint entity linking with deep reinforcement learning. Proc World Wide Web Conf, p.438–447. https://doi.org/10.1145/3308558.3313517
Francis-Landau M, Durrett G, Klein D, 2016. Capturing semantic similarity for entity linking with convolutional neural networks. Proc Conf of the North American
Chapter of the Association for Computational Linguistics: Human Language Technologies, p.1256–1261. https://doi.org/10.18653/v1/N16-1150
Ganea OE, Hofmann T, 2017. Deep joint entity disambiguation with local neural attention. Proc Conf on Empirical Methods in Natural Language Processing, p.2619–2629. https://doi.org/10.18653/v1/D17-1277
Globerson A, Lazic N, Chakrabarti S, et al., 2016. Collective entity resolution with multi-focal attention. Proc 54th Annual Meeting of the Association for Computational Linguistics, p.621–631. https://doi.org/10.18653/v1/P16-1059
Guo ZC, Barbosa D, 2018. Robust named entity disambiguation with random walks. Semant Web, 9(4):459–479. https://doi.org/10.3233/SW-170273
Han XP, Sun L, 2011. A generative entity-mention model for linking entities with knowledge base. Proc 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, p.945–954.
Hoffart J, Yosef MA, Bordino I, et al., 2011. Robust disambiguation of named entities in text. Proc Conf on Empirical Methods in Natural Language Processing, p.782–792.
Hoffmann R, Zhang CL, Ling X, et al., 2011. Knowledge-based weak supervision for information extraction of overlapping relations. Proc 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, p.541–550.
Honnibal M, Dale R, 2009. DAMSEL: the DSTO/Macquarie system for entity-linking. Proc 2nd Text Analysis Conf.
Hou F, Wang RL, He J, et al., 2020. Improving entity linking through semantic reinforced entity embeddings. Proc 58th Annual Meeting of the Association for Computational Linguistics, p.6843–6848. https://doi.org/10.18653/v1/2020.acl-main.612
Le P, Titov I, 2018. Improving entity linking by modeling latent relations between mentions. Proc 56th Annual Meeting of the Association for Computational Linguistics, p.1595–1604. https://doi.org/10.18653/v1/P18-1148
Li ZZ, Feng DW, Li DS, et al., 2020. Learning to select pseudo labels: a semi-supervised method for named entity recognition. Front Inform Technol Electron Eng, 21(6):903–916. https://doi.org/10.1631/FITEE.1800743
Luo G, Huang XJ, Lin CY, et al., 2015. Joint entity recognition and disambiguation. Proc Conf on Empirical Methods in Natural Language Processing, p.879–888. https://doi.org/10.18653/v1/D15-1104
Martins PH, Marinho Z, Martins AFT, 2019. Joint learning of named entity recognition and entity linking. Proc 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop, p.190–196. https://doi.org/10.18653/v1/P19-2026
Medelyan O, Milne D, Legg C, et al., 2009. Mining meaning from Wikipedia. Int J Hum-Comput Stud, 67(9):716–754. https://doi.org/10.1016/j.ijhcs.2009.05.004
Milne D, Witten IH, 2008. Learning to link with Wikipedia. Proc 17th ACM Conf on Information and Knowledge Management, p.509–518. https://doi.org/10.1145/1458082.1458150
Mulang IO, Singh K, Prabhu C, et al., 2020. Evaluating the impact of knowledge graph context on entity disambiguation models. Proc 29th ACM Int Conf on Information & Knowledge Management, p.2157–2160. https://doi.org/10.1145/3340531.3412159
Nguyen DB, Theobald M, Weikum G, 2016. J-NERD: joint named entity recognition and disambiguation with rich linguistic features. Trans Assoc Comput Linguist, 4:215–229. https://doi.org/10.1162/tacl_a_00094
Pennington J, Socher R, Manning C, 2014. GloVe: global vectors for word representation. Proc Conf on Empirical Methods in Natural Language Processing, p.1532–1543. https://doi.org/10.3115/v1/D14-1162
Pershina M, He YF, Grishman R, 2015. Personalized page rank for named entity disambiguation. Proc Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, p.238–243. https://doi.org/10.3115/v1/N15-1026
Runge A, Hovy E, 2020. Exploring neural entity representations for semantic information. Proc 3rd BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, p.204–216. https://doi.org/10.18653/v1/2020.blackboxnlp-1.20
Sevgili O, Shelmanov A, Arkhipov M, et al., 2020. Neural entity linking: a survey of models based on deep learning. https://arxiv.org/abs/2006.00575
Shen W, Wang JY, Han JW, 2015. Entity linking with a knowledge base: issues, techniques, and solutions. IEEE Trans Knowl Data Eng, 27(2):443–460. https://doi.org/10.1109/TKDE.2014.2327028
Vaswani A, Shazeer N, Parmar N, et al., 2017. Attention is all you need. Proc 31st Int Conf on Neural Information Processing Systems, p.6000–6010.
Wang HF, Liu ZQ, 2019. An error recognition method for power equipment defect records based on knowledge graph technology. Front Inform Technol Electron Eng, 20(11):1564–1577. https://doi.org/10.1631/FITEE.1800260
Wang Z, Zhang JW, Feng JL, et al., 2014. Knowledge graph and text jointly embedding. Proc Conf on Empirical Methods in Natural Language Processing, p.1591–1601. https://doi.org/10.3115/v1/D14-1167
Wu JS, Zhang RC, Mao YY, et al., 2020. Dynamic graph convolutional networks for entity linking. Proc Web Conf, p.1149–1159. https://doi.org/10.1145/3366423.3380192
Xin KX, Hua W, Liu Y, et al., 2019. Entity disambiguation based on parse tree neighbours on graph attention network. Proc 20th Int Conf on Web Information Systems Engineering, p.523–537. https://doi.org/10.1007/978-3-030-34223-4_33
Xu P, Barbosa D, 2018. Neural fine-grained entity type classification with hierarchy-aware loss. Proc Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, p.16–25. https://doi.org/10.18653/v1/N18-1002
Xue MG, Cai WM, Su JS, et al., 2019. Neural collective entity linking based on recurrent random walk network learning. Proc 28th Int Joint Conf on Artificial Intelligence, p.5327–5333. https://doi.org/10.24963/ijcai.2019/740
Yamada I, Shindo H, Takeda H, et al., 2016. Joint learning of the embedding of words and entities for named entity disambiguation. Proc 20th SIGNLL Conf on Computational Natural Language Learning, p.250–259. https://doi.org/10.18653/v1/K16-1025
Yang XY, Gu XT, Lin S, et al., 2019. Learning dynamic context augmentation for global entity linking. Proc Conf on Empirical Methods in Natural Language Processing and the 9th Int Joint Conf on Natural Language Processing, p.271–281. https://doi.org/10.18653/v1/D19-1026
Zhang HB, Chen Q, Zhang WW, et al., 2022. HSIE: improving named entity disambiguation with hidden semantic information extractor. Proc 14th Int Conf on Machine Learning and Computing, p.251–257. https://doi.org/10.1145/3529836.3529920
Zhong HP, Zhang JW, Wang Z, et al., 2015. Aligning knowledge and text embeddings by entity descriptions. Proc Conf on Empirical Methods in Natural Language Processing, p.267–272. https://doi.org/10.18653/v1/D15-1031
Author information
Authors and Affiliations
Corresponding author
Additional information
Project supported by the Key-Area Research and Development Program of Guangdong Province, China (No. 2019B010153002), the Program of Marine Economy Development (Six Marine Industries) Special Foundation of Department of Natural Resources of Guangdong Province, China (No. GDNRC [2020]056), the National Natural Science Foundation of China (No. 62002071), the Top Youth Talent Project of Zhujiang Talent Program, China (No. 2019QN01X516), and the Guangdong Provincial Key Laboratory of Cyber-Physical System, China (No. 2020B1212060069)
Contributors
Hongbin ZHANG designed the research. Hongbin ZHANG, Quan CHEN, and Weiwen ZHANG processed the data. Weiwen ZHANG validated the research. Hongbin ZHANG drafted the paper. Quan CHEN helped organize the paper. Hongbin ZHANG and Weiwen ZHANG revised and finalized the paper.
Compliance with ethics guidelines
Hongbin ZHANG, Quan CHEN, and Weiwen ZHANG declare that they have no conflict of interest.
Rights and permissions
About this article
Cite this article
Zhang, H., Chen, Q. & Zhang, W. Improving entity linking with two adaptive features. Front Inform Technol Electron Eng 23, 1620–1630 (2022). https://doi.org/10.1631/FITEE.2100495
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1631/FITEE.2100495