使用TB-LSTM-CRF提高工业中文文本实体识别任务
An Improved Chinese Named Entity Recognition Method with TB-LSTM-CRF
DOI: 10.12677/CSA.2021.113074, PDF,    国家自然科学基金支持
作者: 李嘉正, 周佳乐, 程良伦：广东工业大学 自动化学院，广东 广州
关键词: 中文自然语言处理；中文命名实体识别；知识图谱；Chinese Natural Language Processing； Chinese Named Entity Recognition； Knowledge Graph
摘要: 由于缺乏自然定界符，中文命名实体识别(NER)比英文更具挑战性。虽然中文分词(CWS)通常被认为是中文NER的关键和开放问题，但其准确性对于下游模型训练至关重要，而且经常遭受语音不足(OOV)的困扰。本文提出了一种改进的中文NER模型，称为TB-LSTM-CRF，该模型在LSTM-CRF之上引入了一个变压器块。带有Transformer Block的拟议模型利用self-attention机制从相邻字符和句子上下文中捕获信息。同时本文使用了一个全新的工业场景数据集，在此数据集上，与使用LSTM-CRF的基线相比，实验结果表明，TB-LSTM-CRF方法具有竞争力，几乎不需要任何外部资源，例如参数迁移。
Abstract: Owing to the lack of natural delimiters, Chinese Named Entity Recognition (NER) is more challenging than it in English. While Chinese Word Segmentation (CWS) is generally regarded as key and open problem for Chinese NER, its accuracy is critical for the downstream models trainings and it also often suffers from Out-of-Vocabulary (OOV). In this paper, we propose an improved Chinese NER model called TB-LSTM-CRF, which introduces a Transformer Block on top of LSTM- CRF. The proposed model with Transformer Block exploits the self-attention mechanism to capture the information from adjacent characters and sentence contexts. At the same time, this article uses a brand new industrial scene data set. On this data set, with the LSTM-CRF scale, the experimental results show that the TB-LSTM-CRF method is competitive and hardly requires any external resources, such as parameter migration.
文章引用：李嘉正, 周佳乐, 程良伦. 使用TB-LSTM-CRF提高工业中文文本实体识别任务[J]. 计算机科学与应用, 2021, 11(3): 720-728. https://doi.org/10.12677/CSA.2021.113074

参考文献

[1] Yadav, V. and Bethard, S. (2019) A Survey on Recent Advances in Named Entity Recognition from Deep Learning Models. arXiv:1910.11470.
[2] Liu, L., Shang, J., Ren, X., Xu, F.F., Gui, H., Peng, J. and Han, J. (2018) Empower Sequence Labeling with Task- Aware Neural Language Model. 32nd AAAI Conference on Artificial Intelligence, New Orleans, 2-7 February 2018, 5253-5260.
[3] Yao, L., Ge, J., Li, C., Yao, Y., Li, Z., Zeng, J., et al. (2019) Word Seg-mentation for Chinese Judicial Documents. International Conference of Pioneering Computer Scientists, Engineers and Educators, 20-23 September 2019, Guilin, 466-478. [Google Scholar] [CrossRef] 
[4] Chen, X., Shi, Z., Qiu, X. and Huang, X. (2017) Adversarial Multi-Criteria Learning for Chinese Word Segmentation. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, Vancouver, 30 July-4 August, 1193-1203. [Google Scholar] [CrossRef] 
[5] Li, H., Hagiwara, M., Li, Q. and Ji, H. (2014). Comparison of the Im-pact of Word Segmentation on Name Tagging for Chinese and Japanese. Proceedings of the 9th International Conference on Language Resources and Evaluation, Reykjavik, 26-31 May 2014, 2532-2536.
[6] Yang, Y. and Katiyar, A. (2020) Simple Few-Shot Named Entity Recognition with Structured Nearest Neighbor Learning. Proceedings of the 2020 Con-ference on Empirical Methods in Natural Language Processing (EMNLP), Online, November 2020, 6365-6375. [Google Scholar] [CrossRef] 
[7] Al-Smadi, M., Al-Zboon, S., Jararweh, Y. and Juola, P. (2020) Transfer Learning for Arabic Named Entity Recognition with Deep Neural Networks. IEEE Access, 8, 37736-37745. [Google Scholar] [CrossRef] 
[8] Popovski, G., Seljak, B.K. and Eftimov, T. (2020) A Survey of Named-Entity Recognition Methods for Food Information Extraction. IEEE Access, 8, 31586-31594. [Google Scholar] [CrossRef] 
[9] Prakash, R. and Dubey, R.K. (2020) Transfer Learning and Domain Adaptation for Named-Entity Recognition. In: Johri, P., Verma, J. and Paul, S., Eds., Applications of Machine Learning, Springer, Singapore, 67-73. [Google Scholar] [CrossRef] 
[10] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., et al. (2017) Attention Is All You Need. 31st Conference on Neural Information Processing Systems, Long Beach, 4-9 December 2017, 5998-6008.
[11] Weischedel, R., Pradhan, S., Ramshaw, L., Palmer, M., Xue, N., Marcus, M., et al. (2011) OntoNotes Release 4.0. LDC2011T03. Linguistic Data Consortium, Philadelphia.
[12] Peng, N. and Dredze, M. (2015) Named Entity Recognition for Chinese Social Media with Jointly Trained Embeddings. Pro-ceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, Lisbon, September 2015 548-554. [Google Scholar] [CrossRef] 
[13] Zhang, Y. and Yang, J. (2018) Chinese NER Using Lattice LSTM. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, July 2018, 1554-1564. [Google Scholar] [CrossRef] 
[14] Ratinov, L. and Roth, D. (2009) Design Challenges and Misconceptions in Named Entity Recognition. Proceedings of the Thirteenth Conference on Computational Natural Language Learning, June 2009, 147-155. [Google Scholar] [CrossRef] 
[15] Geng, X., Wang, L., Wang, X., et al. (2020) How Does Selective Mechanism Improve Self-Attention Networks? Proceedings of the 58th Annual Meeting of the Association for Computa-tional Linguistics, July 2020, 2986-2995. [Google Scholar] [CrossRef] 
[16] Zhu, Y. and Wang, G. (2019) CAN-NER: Convolutional At-tention Network for Chinese Named Entity Recognition. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Vol. 1 (Long and Short Pa-pers), Minneapolis, 2-7 June 2019, 3384-3393.
[17] Yang, J., Teng, Z., Zhang, M. and Zhang, Y. (2016) Combining Discrete and Neural Features for Sequence Labeling. In: Gelbukh, A., Ed., Computational Linguistics and Intelligent Text Processing, Springer, Cham, 140-154. [Google Scholar] [CrossRef] 
[18] Peng, N. and Dredze, M. (2016) Improving Named Entity Recognition for Chinese Social Media with Word Segmentation Representation Learning. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Berlin, August 2016, 149-155. [Google Scholar] [CrossRef]