基于深度学习的命名实体识别算法
Named Entity Recognition Algorithm Based on Deep Learning
DOI: 10.12677/CSA.2021.113064, PDF,  被引量    国家自然科学基金支持
作者: 陈 娟：广东省信息物理融合系统重点实验室，广东 广州；王卓薇：广东工业大学，广东 广州；程良伦：广东工业大学计算机学院，广东 广州
关键词: 知识图谱；深度学习；实体识别；Knowledge Graph； Deep Learning； Entity Recognition
摘要: 命名实体识别(Named Entity Recognition, NER)的定义是从自由文本中识别出属于预定义类别的文本片段(如人名、地理位置名、机构组织名等)。命名实体识别一直是许多自然语言应用的基础，例如问题回答、提取文本摘要和知识库建立。早期的NER系统在实现良好性能方面取得了巨大的成功，其代价是人类工程在设计特定领域的特征和规则方面付出的代价。近年来，非线性处理的连续实值向量表示和语义组合使得深度学习在命名实体识别系统中发挥了很好的作用。在本文中，我们提供了一种基于深度学习的命名实体识别算法。首先我们随机初始化训练集中的每个字特征，并在获取该字典句子中每个字的特征之后，利用周期卷积来得到其固定长度的特征，以此作为句子特征;随后训练数据自动编码器，通过栈式自动编码器得到高层句子的特征;最后通过高层句特征与字特征的组合训练字的标注网络模型来得到未知字的标注值，再进行实体扩展(分类，属性，副标题)，最后利用马尔科夫逻辑网络优化整体识别效果。
Abstract: Named entity recognition is the task of identifying rigid indicators from text belonging to predefined semantic types (such as person, location, organizations, and so on). NER has been the basis for many natural language applications, such as question answering, text summarization and machine translation. Early NER systems had great success in achieving good performance at the cost of human engineering in designing domain-specific features and rules. In recent years, deep learning has been used in NER systems through continuous real-valued vector representations and semantic combinations of nonlinear processing, resulting in the most advanced performance. In this article, we provide an entity recognition technique based on deep learning. Firstly, each word feature in the training set is randomly initialized, and the feature of each word in the sentence is obtained based on the dictionary. Then, the fixed-length feature is obtained by periodic convolution of different length of sentence features, which are used as sentence features. Then the data autoencoder is trained to get the features of high-level sentences through the stack autoencoder. Finally, the combination of high level sentence features and word features is used to train the annotation network model of words, and the annotation value of unknown words is obtained based on the annotation model, and then the entity expansion (classification, attribute, subtitle) is carried out. Finally, the overall recognition effect is optimized by using Markov logic network.
文章引用：陈娟, 王卓薇, 程良伦. 基于深度学习的命名实体识别算法[J]. 计算机科学与应用, 2021, 11(3): 628-634. https://doi.org/10.12677/CSA.2021.113064

参考文献

[1] Christian, B., Heath, T. and Berners-Lee, T. (2009) Linked Data—The Story So Far. International Journal on Semantic Web and Information Systems, 5, 1-22. [Google Scholar] [CrossRef] 
[2] Bollacker, K., Cook, R. and Tufts, P. (2007) Freebase: A Shared Database of Structured General Human Knowledge. AAAI, 7, 1962-1963.
[3] Nadeau, D. and Sekine, S. (2007) A Survey of Named Entity Recognition and Classification. Lingvisticæ Investigationes, 30, 3-26. [Google Scholar] [CrossRef] 
[4] Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K. and Kuksa, P. (2011) Natural Language Processing (Almost) from Scratch. Journal of Machine Learning Research, 12, 2493-2537.
[5] Huang, Z., Xu, W. and Yu, K. (2015) Bidirectional LSTM-CRF Models for Sequence Tagging. arXiv preprint arXiv:1508.01991.
[6] Lample, G., Ballesteros, M., Subramanian, S., Kawakami, K. and Dyer, C. (2016) Neural Architectures for Named Entity Recognition. Proceedings of the 2016 Con-ference of the North American Chapter of the Association for Computational Linguistics: Human Language Technolo-gies, San Diego, June 2016, 260-270. [Google Scholar] [CrossRef] 
[7] Chiu, J.P. and Nichols, E. (2016) Named Entity Recognition with Bidi-rectional LSTM-CNNs. Transactions of the Association for Computational Linguistics, 4, 357-370. [Google Scholar] [CrossRef] 
[8] Peters, M.E., Ammar, W., Bhagavatula, C. and Power, R. (2017) Semi-Supervised Sequence Tagging with Bidirectional Language Models. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 1, 1756-1765. [Google Scholar] [CrossRef] 
[9] Patawar, M.L. and Potey, M. (2015) Approaches to Named Entity Recognition: A Survey. International Journal of Innovative Research in Computer and Communication Engineering, 3, 12201-12208.
[10] Saju, C.J. and Shaja, A. (2017) A Survey on Ef-ficient Extraction of Named Entities from New Domains Using Big Data Analytics. 2017 Second International Confer-ence on Recent Trends and Challenges in Computational Models (ICRTCCM), Tindivanam, 3-4 February 2017, 170-175. [Google Scholar] [CrossRef] 
[11] Dai, X. (2018) Recognizing Complex Entity Mentions: A Review and Future Directions. Proceedings of ACL 2018, Student Research Workshop, Melbourne, July 2018, 37-44. [Google Scholar] [CrossRef] 
[12] Yadav, V. and Bethard, S. (2018) A Survey on Recent Advances in Named Entity Recognition from Deep Learning Models. Proceedings of the 27th International Conference on Computa-tional Linguistics, Santa Fe, 20-26 August 2018, 2145-2158.
[13] Goyal, A., Gupta, V. and Kumar, M. (2018) Recent Named Entity Recognition and Classification Techniques: A Systematic Review. Computer Science Review, 29, 21-43. [Google Scholar] [CrossRef] 
[14] Rau, L.F. (1991) Extracting Company Names from Text. Pro-ceedings of the 7th IEEE Conference on Artificial Intelligence Applications, Miami Beach, 24-28 February 1991, 29-32.
[15] Grishman, R. and Sundheim, B. (1996) Message Understanding Conference-6: A Brief History. Proceed-ings of the 16th International Conference on Computational Linguistics, 1, 466-471. [Google Scholar] [CrossRef] 
[16] Chinchor, N.A. (1998) Overview of MUC-7/MET-2. Proceedings of the 7th Message Understanding Conference.
[17] 孙茂松, 黄昌宁, 高海燕, 等. 中文姓名的自动辨识[J]. 中文信息学报, 1995, 9(2): 16-27.
[18] 张小衡, 王玲玲. 中文机构名称的识别与分析[J]. 中文信息学报, 1997, 11(4): 21-32.
[19] Zhang, Y. and Zhou, J.F. (2000) A Trainable Method for Extracting Chinese Entity Names and Their Rela-tions. In: Proceedings of the 2nd Chinese Language Processing Workshop, Hong Kong, October 2000, 66-76. [Google Scholar] [CrossRef] 
[20] Bikel, D.M., Schwarta, R. and Weischedel, R.M. (1999) An Algo-rithm that Learns What’s in a Name. Machine Learning, 34, 211-231. [Google Scholar] [CrossRef] 
[21] Liao, W. and Veeramachaneni, S. (2009) A Simple Semi-supervised Algorithm for Named Entity Recognition. Proceedings of the NAACL HLT 2009 Workshop on Semi-Supervised Learning for Natural Language Processing, Boulder, June 2009, 58-65. [Google Scholar] [CrossRef] 
[22] Ratinov, L. and Roth, D. (2009) Design Challenges and Miscon-ceptions in Named Entity Recognition. Proceedings of the 13th Conference on Computational Natural Language Learn-ing, Boulder, 4-5 June 2009, 147-155. [Google Scholar] [CrossRef] 
[23] 冯元勇, 孙乐, 李文波, 等. 基于单字提示特征的中文命名实体识别快速算法[J]. 中文信息学报, 2008, 22(1): 105-110.
[24] 郑逢强, 林磊, 刘秉权, 等. 《知网》在命名实体识别中的应用研究[J]. 中文信息学报, 2008, 22(5): 97-101.
[25] Chen, X., Qiu, X., Zhu, C., Liu, P. and Huang, X. (2015) Long Short-Term Memory Neural Networks for Chinese Word Segmentation. Proceedings of the 2015 Confer-ence on Empirical Methods in Natural Language Processing, Lisbon, September 2015, 1197-1206. http://aclweb.org/anthology/D15-1141 [Google Scholar] [CrossRef] 
[26] Limsopatham, N. and Collier, N. (2016) Bidirectional LSTM for Named Entity Recognition in Twitter Messages. Proceedings of the 2nd Workshop on Noisy User-generated Text (WNUT), Osaka, 11 December 2016, 145-152.
[27] Dong, C., Zhang, J., Zong, C., Hattori, M. and Di, H. (2016) Character-Based LSTM-CRF with Radical-Level Features for Chinese Named Entity Recognition. In: Lin, C.Y., Xue, N., Zhao, D., Huang, X. and Feng, Y., Eds., Natural Language Understanding and Intelligent Applications, Springer, Cham, 239-250. [Google Scholar] [CrossRef] 
[28] Zhang, Y. and Yang, J. (2018) Chinese NER Using Lattice LSTM. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 1, 1554-1564. [Google Scholar] [CrossRef] 
[29] Ee, S. and Xiang, Y. (2017) Chinese Named Entity Recognition with Character-Word Mixed Embedding. Proceedings of the 2017 ACM on Conference on Information and Knowledge Man-agement, ACM, 2055-2058.