关系重叠是实体关系抽取任务中的一个难题,具体表现为文本中的一个实体存在于多个关系之中。法律文书作为一类存在大量关系重叠的文本,在对其进行关系抽取时要识别这些重叠的关系,而现有的关系抽取方法存在错误传播或识别率不高的问题。针对这种情况,该文提出了联合标注法,构建了联合抽取模型。模型在编码器中获取文本中的关系信息,使用拼接法或权值法来对关系信息和文本信息进行融合,使编码带有关系信息,采用共享的解码器对重叠关系进行识别。在法律文书数据集上进行实验,结果表明,与现有前沿关系抽取方法相比,该关系抽取方法能够更好地识别重叠关系。
Abstract
Overlapping relation, i.e. one entity with multiple relations in a text, is a challenging issue in the task of entities relation extraction. Legal documents, as a type of text with a large number of overlapping relations, typically demonstrate the error propagation or low recognition rate owing to this issue for current relation extraction methods. This paper presents a joint extraction model based on a joint annotation method. Our model obtains the relation information in the text encoder, and embeds the relation information with the text representation via concatenating or weighting. Then a shared decoder is employed to identify the overlapping relations. Experimental results show that, compared with the existing advanced relation extraction methods, this proposed method can better identify overlapping relation.
关键词
联合抽取 /
关系抽取 /
关系重叠
{{custom_keyword}} /
Key words
joint extraction /
relation extraction /
overlapping relation
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] LUO B, FENG Y, XU J, et al. Learning to predict charges for criminal cases with legal basis[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Copenhagen, 2017: 2727-2736.
[2] HUZ K, LI X, TU C C, et al. Few-shot charge prediction with discriminative legal attribute[C]//Proceedings of the 27th International Conference on Computational Linguistics. Santa Fe, 2018:1-12.
[3] 何庆,汤庸,黄永钊. 基于本体的法律知识库的研究与实现[J]. 计算机科学, 2007, 34(2): 175-177.
[4] 黄菡,王宏宇,王晓光. 结合主动学习的条件随机场模型用于法律术语的自动识别[J]. 数据分析与知识发现,2019,3(06): 66-74.
[5] 高丹,彭敦陆,刘丛. 海量法律文书中基于CNN的实体关系抽取技术[J]. 小型微型计算机系统, 2018, 39(05): 1021-1026.
[6] 张海楠,伍大勇,刘悦,等. 基于深度神经网络的中文命名实体识别[J]. 中文信息学报, 2017, 31(04): 28-35.
[7] HUANG F, AHUJA A, DOWNEY D, et al. Learning representations for weakly supervised natural language processing tasks[J]. Computational Linguistics, 2014, 40(01): 85-120.
[8] ZHAI F, POTDAR S, XIANG B, et al. Neural models for sequence chunking[C]//Proceedings of the AAAI International Conference, 2017: 3365-3371.
[9] ZHENG S, WANG F, BAO H, et al. Joint extraction of entities and relations based on a novel tagging scheme[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017:1227-1236.
[10] WEI Z, SU J, WANG Y, et al. A novel cascade binary tagging framework for relational triple extraction[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 1476-1488.
[11] 俞鸿魁,张华平,刘群,等. 基于层叠隐马尔可夫模型的中文命名实体识别[J]. 通信学报, 2006, 27(02): 87-94.
[12] PASSOS A, KUMAR V, MCCALLUM A. Lexicon infused phrase embeddings for named entity resolution[C]//Proceedings of the International Conference on Computation Linguistics, 2014: 78-86.
[13] CHIU J P, NICHOLS E. Named entity recognition with bidirectional LSTM-CNNS[C]//Processdings of Transactions of the Association for Computational Linguistics, 2016: 357-370.
[14] LAMPLE G, BALLESTEROS M, SUBRAMANIAN S, et al. Neural architectures for named entity recognition[C]//Proceedings of the NAACL International Conference, 2016: 260-270.
[15] 成于思,施云涛. 融合词典特征的Bi-LSTM-WCRF中文人名识别[J]. 中文信息学报,2020,34(04): 69-76.
[16] LIU C Y, SUN W B, CHAO W H, et al. Convolution neural network for relation extraction[C]//Proceedings of the International Conference on Advanced Data Mining and Applications, Part Ⅱ, 2013: 231-242.
[17] ZENG D, LIU K, LAI S, et al. Relation classification via convolutional deep neural network[C]//Proceedings of the 25th International Conference on Computational Linguistics. Dublin, Ireland, 2014: 2335-2344.
[18] SANTOS C,XIANG B,ZHOU B. Classifying relations by ranking with convolutional neural networks[C]//Proceedings of the 53th ACL International Conference, 2015: 626-634.
[19] MIWA M, BANSAL M. End-to-end relation extraction using LSTMs on sequences and tree structures[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics,2016: 1105- 1116.
[20] ZENG X, ZENG D, HU S, et al. Extracting relational facts by an end-to-end neural model with copy mechanism[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018: 506-514.
[21] LI X, YIN F, SUN Z, et al. Entity-relation extraction as multi-turn question answering[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019: 1340-1350.
[22] XUE K, ZHOU Y, MA Z, et al. Fine-tuning BERT for joint entity and relation extraction in Chinese medical text[C]//Proceedings of the 2019 IEEE International Conference on Bioinformatics and Biomedicine, 2019: 892-897.
[23] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the Advances in Neural Information Processing Systems, 2017: 6000-6010.
[24] DEVLIN J, ZHANG M, LEE K, et al. BERT: Pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019: 4171-4186.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
十三五国家重点研发计划课题(2017YFB0503701);湖北省技术创新专项(2019AAA045)
{{custom_fund}}
PDF(1928 KB)
