微信服务号

微信订阅号

2025年4月4日 10:13 星期五
首页 > 过刊浏览>2018年第29卷第8期 >2427-2447. DOI:10.13328/j.cnki.jos.005485
PDF HTML阅读 XML下载 导出引用 引用提醒
基于双向LSTM网络的不确定和否定作用范围识别
作者:
作者简介:

钱忠(1989-),男,江苏常熟人,博士,主要研究领域为自然语言处理;周国栋(1967-),男,博士,教授,博士生导师,CCF高级会员,主要研究领域为自然语言处理;李培峰(1971-),男,博士,教授,博士生导师,CCF高级会员,主要研究领域为自然语言处理;朱巧明(1963-),男,博士,教授,博士生导师,CCF杰出会员,主要研究领域为自然语言处理

通讯作者:

钱忠,E-mail:qianzhongqz@163.com;李培峰,E-mail:pfli@suda.edu.cn

基金项目:

国家自然科学基金（61331011，61472265，61772354）；江苏省科技计划（BK20151222）


Speculation and Negation Scope Detection via Bidirectional LSTM Neural Networks
Author:
Fund Project:

National Natural Science Foundation of China (61331011, 61472265, 61772354); Science and Technology Project of Jiangsu Province (BK20151222)

  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献 [56]
    • |
  • 相似文献 [20]
    • | | |
  • 文章评论
    摘要:

    不确定和否定信息抽取，是自然语言处理领域中的重要任务和研究热点.针对不确定和否定作用范围识别任务，提出一种基于两层双向LSTM神经网络的作用范围识别方法.首先，对于从线索词到达词语的句法路径，使用第1层双向LSTM神经网络从中学习到有用特征；接着，将词法特征与句法路径特征一起组成当前词语的特征表示；最后，将作用范围识别问题看作序列标注任务，利用第2层双向LSTM神经网络界定当前线索词的作用范围.实验结果表明，所提出的模型优于其他神经网络模型，并在BioScope生物医学语料上取得了良好性能.其中，在Abstracts子语料上的不确定和否定作用范围识别精确率分别达到86.20%和80.28%.

    Abstract:

    Speculation and negation information extraction is an important task and research focus in natural language processing (NLP). This paper proposes a two-layer bidirectional long short-term memory (LSTM) neural network model for speculation and negation scope detection. Firstly, a bidirectional LSTM neural network is utilized in the first layer to learn useful feature representations from the syntactic path which is from the cue to the token. Then, lexical features and syntactic path features are concatenated into the feature representations of the token. Finally, taking the scope detection problem as a sequence labeling task, another bidirectional LSTM neural network is employed in the second layer to identify the scope of the current cue. The experimental results show that the presented model is superior to other neural network models and attains excellent performances on BioScope corpus. Particularly, the model achieves the accuracy (percentage of correct scopes) of 86.20% and 80.28% on speculation and negation scope detection on Abstracts subcorpus, respectively.

    参考文献
    [1] Lakoff G. Hedges:A study in meaning criteria and the logic of fuzzy concepts. Journal of Philosophical Logic, 1973,2(4):458-508.[doi:10.1007/BF00262952]
    [2] Morante R, Sporleder C. Modality and negation:An introduction to the special issue. Computational Linguistics, 2012,38(2):223-260.[doi:10.1162/COLI_a_00095]
    [3] Pustejovsky J, Knippen B, Littman J, Saurí R. Temporal and event information in natural language text. Language Resources and Evaluation, 2005,39(2):123-164.[doi:10.1145/1314493.1314495]
    [4] Angeli G, Nayak N, Manning CD. Combining natural logic and shallow reasoning for question answering. In:Erk K, Smith NA, eds. Proc. of the 2016 Conf. on the 54th Annual Meeting of the Association for Computational Linguistics (Vol.1:Long Papers). Stroudsburg:Association for Computational Linguistics, 2016. 442-452.
    [5] Wiebe J, Wilson T, Cardie C. Annotating expressions of opinions and emotions in language. Language Resources & Evaluation, 2005,39(2-3):165-210.[doi:10.1007/s10579-005-7880-9]
    [6] de Marneffe MC, Maccartney B, Grenager T, Cer D, Rafferty A, Manning CD. Learning to distinguish valid textual entailments. In Proc. of the 2nd Pascal Challenges Workshop on Recognising Textual Entailment (RTE-2). 2006.
    [7] Hickl A, Bensley J. A discourse commitment-based framework for recognizing textual entailment. In:Dagan I, Dolan B, Giampiccolo D, Magnini B, eds. Proc. of the Workshop on Textual Entailment and Paraphrasing. Stroudsburg:Association for Computational Linguistics, 2007. 171-176.
    [8] Saurí R, Pustejovsky J. Determining modality and factuality for text entailment. In:Proc. of the 1st IEEE Int'l Conf. on Semantic Computing (ICSC 2007). 2007. 509-516.[doi:10.1109/ICSC.2007.80]
    [9] Councill IG, McDonald R, Velikovich L. What's great and what's not:Learning to classify the scope of negation for improved sentiment analysis. In:Morante R, Sporleder C, eds. Proc. of the Workshop on Negation and Speculation in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2010. 51-59.
    [10] Saurí R. A factuality profiler for eventualities in text.[Ph.D. Thesis]. Waltham:Brandeis University, 2008.
    [11] Saurí R, Pustejovsky J. Are you sure that this happened? Assessing the factuality degree of events in text. Computational Linguistics, 2012,38(2):261-299.[doi:10.1162/COLI_a_00096]
    [12] Tang BZ, Wang XL, Wang X, Yuan B, Fan FX. A cascade method for detecting hedges and their scope in natural language text. In:Farkas R, Vincze V, Szarvas G, Móra G, Csirik J, eds. Proc. of the 14th Conf. on Computational Natural Language Learning-Shared Task. Stroudsburg:Association for Computational Linguistics, 2010. 13-17.
    [13] Vincze V, Szarvas G, Farkas R, Móra G, Csirik J. The BioScope corpus:Biomedical texts annotated for uncertainty, negation and their scopes. BMC Bioinformatics, 2008,9(Suppl.11):1-9.[doi:10.1186/1471-2105-9-S11-S9]
    [14] Özgür A, Radev DR. Detecting speculations and their scopes in scientific text. In:Koehn P, Mihalcea R, eds. Proc. of the 2009 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2009. 1398-1407.
    [15] Øvrelid L, Velldal E, Oepen S. Syntactic scope resolution in uncertainty analysis. In:Huang CR, Jurafsky D, eds. Proc. of the 23rd Int'l Conf. on Computational Linguistics. Stroudsburg:Association for Computational Linguistics, 2010. 1379-1387.
    [16] Zou BW, Zhou GD, Zhu QM. Tree kernel-based negation and speculation scope detection with structured syntactic parse features. In:Baldwin T, Korhonen A, eds. Proc. of the 2013 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2013. 968-976.
    [17] Li JH, Zhou GD, Wang HL, Zhu QM. Learning the scope of negation via shallow semantic parsing. In:Huang CR, Jurafsky D, eds. Proc. of the 23rd Int'l Conf. on Computational Linguistics. Stroudsburg:Association for Computational Linguistics, 2010. 671-679.
    [18] Velldal E, Øvrelid L, Read J, Oepen S. Speculation and negation:Rules, rankers, and the role of syntax. Computational Linguistics, 2010,38(2):369-410.[doi:10.1162/COLI_a_00126]
    [19] Roth M, Lapata M. Neural semantic role labeling with dependency path embeddings. In:Erk K, Smith NA, eds. Proc. of the 54th Annual Meeting of the Association for Computational Linguistics (Vol.1:Long Papers). Stroudsburg:Association for Computational Linguistics, 2016. 1192-1202.
    [20] Zhou J, Xu W. End-to-End learning of semantic role labeling using recurrent neural networks. In:Zong CQ, Strube M, eds. Proc. of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th Int'l Joint Conf. on Natural Language Processing (Vol.1:Long Papers). Stroudsburg:Association for Computational Linguistics, 2015. 1127-1137.
    [21] Xu JC, Chen DL, Qiu XP, Huang XJ. Cached long short-term memory neural networks for document-level sentiment classification. In:Duh K, Carreras X, eds. Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2016. 1660-1669.
    [22] Wang XY, Jiang WJ, Luo ZY. Combination of convolutional and recurrent neural network for sentiment analysis of short texts. In:Matsumoto Y, Prasad R, eds. Proc. of the 26th Int'l Conf. on Computational Linguistics:Technical Papers (COLING 2016). Stroudsburg:Association for Computational Linguistics, 2016. 2428-2437.
    [23] Chen YB, Xu LH, Liu K, Zeng DJ, Zhao J. Event extraction via dynamic multi-pooling convolutional neural networks. In:Zong CQ, Strube M, eds. Proc. of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th Int'l Joint Conf. on Natural Language Processing (Vol.1:Long Papers). Stroudsburg:Association for Computational Linguistics, 2015. 167-176.
    [24] Nguyen TH, Cho K, Grishman R. Joint event extraction via recurrent neural networks. In:Nenkova A, Rambow O, eds. Proc. of the 2016 Conf. of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies. Stroudsburg:Association for Computational Linguistics, 2016. 300-309.
    [25] Xu Y, Mou LL, Li G, Chen YC, Peng H, Jin Z. Classifying relations via long short term memory networks along shortest dependency paths. In:Callison-Burch C, Su J, eds. Proc. of the 2015 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2015. 1785-1794.
    [26] Cai R, Zhang XD, Wang HF. Bidirectional recurrent convolutional neural network for relation classification. In:Erk K, Smith NA, eds. Proc. of the 54th Annual Meeting of the Association for Computational Linguistics. Stroudsburg:Association for Computational Linguistics, 2016. 756-765.
    [27] Yang YL, Tong YH, Ma SL Deng ZH. A position encoding convolutional neural network based on dependency tree for relation classification. In:Duh K, Carreras X, eds. Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2016. 65-74.
    [28] Zhou J, Cao Y, Wang XG, Li P, Xu W. Deep recurrent models with fast-forward connections for neural machine translation. Trans. of the Association for Computational Linguistics, 2016,4:371-383.
    [29] Wang MX, Lu ZD, Li H, Liu Q. Memory-Enhanced decoder for neural machine translation. In:Duh K, Carreras X, eds. Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2016. 278-286.
    [30] Hochreiter S, Schmidhuber J. Long short-term memory. Neural Computation, 1997,9(8):1735-1780.[doi:10.1162/neco.1997.9.8. 1735]
    [31] Fancellu F, Lopez A, Webber B. Neural networks for negation scope detection. In:Erk K, Smith NA, eds. Proc. of the 54th Annual Meeting of the Association for Computational Linguistics. Stroudsburg:Association for Computational Linguistics, 2016. 495-504.
    [32] Qian Z, Li PF, Zhu QM, Zhou GD, Luo ZC, Luo W. Speculation and negation scope detection via convolutional neural networks. In:Duh K, Carreras X, eds. Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2016. 815-825.
    [33] Chapman WW, Bridewell W, Hanbury P, Cooper GF, Buchanan BG. A simple algorithm for identifying negated findings and diseases in discharge summaries. Journal of Biomedical Informatics, 2001,34(5):301-310.[doi:10.1006/jbin.2001.1029]
    [34] Aronow D, Feng FF, Croft WB. Research paper:Ad hoc classification of radiology reports. Journal of the American Medical Informatics Association, 1999,6(5):393-411.[doi:10.1136/jamia.1999.0060393]
    [35] Boytcheva S, Strupchanska A, Paskaleva E, Tcharaktchiev D. Some aspects of negation processing in electronic health records. In:Paskaleva E, Piperidis S, eds. Proc. of the Int'l Workshop Language and Speech Infrastructure for Information Access in the Balkan Countries Held in Conjunction with RANLP 2005. INCOMA, 2005. 1-8.
    [36] Huang Y, Lowe HJ. Research paper:A novel hybrid approach to automated negation detection in clinical radiology reports. Journal of the American Medical Informatics Association, 2007,14(3):304-311.[doi:10.1197/jamia.M2284]
    [37] Rokach L, Romano R, Maimon O. Negation recognition in medical narrative reports. Information Retrieval Journal, 2008,11(6):499-538.[doi:10.1007/s10791-008-9061-0]
    [38] Morante R, Liekens A, Daelemans W. Learning the scope of negation in biomedical texts. In:Lapata M, Ng HT, eds. Proc. of the 2008 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2008. 715-724.
    [39] Morante R, Daelemans W. Learning the scope of hedge cues in biomedical texts. In Proc. of the BioNLP 2009 Workshop. Stroudsburg:Association for Computational Linguistics, 2009. 28-36.
    [40] Morante R, Daelemans W. A metalearning approach to processing the scope of negation. In Proc. of the 13th Conf. on Computational Natural Language Learning (CoNLL 2009). Stroudsburg:Association for Computational Linguistics, 2009. 21-29.
    [41] Zhu QM, Li JH, Wang HL, Zhou GD. A unified framework for scope learning via simplified shallow semantic parsing. In:Li H, Màrquez L, eds. Proc. of the 2010 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2010. 714-724.
    [42] Zou BW, Qian Z, Chen ZC, Zhu QM, Zhou GD. Negation and uncertainty information extraction oriented to natural language text. Ruan Jian Xue Bao/Journal of Software, 2016,27(2):309-328(in Chinese with English abstract). http://www.jos.org.cn/1000-9825/4860.htm[doi:10.13328/j.cnki.jos.004860]
    [43] Cierniak R. A new approach to image reconstruction from projections using a recurrent neural network. Applied Mathematics and Computer Science, 2008,18(2):147-157.[doi:10.2478/v10006-008-0014-y]
    [44] Juang CF, Chiou CT, Huang HJ. Noisy speech recognition by hierarchical recurrent neural fuzzy networks. In:Proc. of Int'l Symp. on Circuits and Systems (ISCAS 2005). IEEE, 2005. 5122-5125.[doi:http://dx.doi.org/10.1109/ISCAS.2005.1465787]
    [45] Lin R, Liu SJ, Yang MY, Li M, Zhou M, Li S. Hierarchical recurrent neural network for document modeling. In:Callison-Burch C, Su J, eds. Proc. of the 2015 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2015. 899-907.
    [46] Liu PF, Qiu XP, Chen XC, Wu SY, Huang XJ. Multi-Timescale long short-term memory neural network for modelling sentences and documents. In:Callison-Burch C, Su J, eds. Proc. of the 2015 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2015. 2326-2335.
    [47] Zhou P, Shi W, Tian J, Qi ZY, Li BC, Hao HW, Xu B. Attention-Based bidirectional long short-term memory networks for relation classification. In:Erk K, Smith NA, eds. Proc. of the 54th Annual Meeting of the Association for Computational Linguistics (Vol.2:Short Papers). Stroudsburg:Association for Computational Linguistics, 2016. 207-212.
    [48] Tang DY, Qin B, Liu T. Document modeling with gated recurrent neural network for sentiment classification. In:Callison-Burch C, Su J, eds. Proc. of the 2015 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2015. 1422-1432.
    [49] Chen HM, Sun MS, Tu CC, Lin YK, Liu ZY. Neural sentiment classification with user and product attention. In:Carreras X, Duh K, eds. Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2016. 1650-1659.
    [50] Lapponi E, Velldal E, Øvrelid L, Read J. UiO2:Sequence-labeling negation using dependency features. In:Agirre E, Bos J, Diab M, eds. Proc. of the 1st Joint Conf. on Lexical and Computational Semantics-Vol.1:Proc. of the Main Conf. and the Shared Task, and Vol.2:Proc. of the 6th Int'l Workshop on Semantic Evaluation. Stroudsburg:Association for Computational Linguistics, 2012. 319-327.
    [51] Zeng DJ, Liu K, Lai SW, Zhou GY, Zhao J. Relation classification via convolutional deep neural network. In:Tsujii J, Hajic J, eds. Proc. of the 25th Int'l Conf. on Computational Linguistics:Technical Papers (COLING 2014). Stroudsburg:Association for Computational Linguistics, 2014. 2335-2344.
    [52] Hinton GE, Srivastava N, Krizhevsky A, Sutskever I, Salakhutdinov R. Improving neural networks by preventing co-adaptation of feature detectors. CoRR, abs/1207.0580, 2012.
    [53] Tateisi Y, Yakushiji A, Ohta T, Tsujii J. Syntax annotation for the GENIA corpus. In:Dale R, Wong KF, Su J, Kwong YO, eds. Proc. of 2nd Int'l Joint Conf. on Natural Language Processing (IJCNLP 2005). Stroudsburg:Association for Computational Linguistics, 2005. 220-225.
    [54] Farkas R, Vincze V, Móra G, Csirik J, Szarvas G. The conll-2010 shared task:learning to detect hedges and their scope in natural language text. In Farkas R, Vincze V, Szarvas G, Móra G, Csirik J, eds. Proc. of the 14th Conf. on Computational Natural Language Learning-Shared Task. Stroudsburg:Association for Computational Linguistics, 2010. 1-12.
    附中文参考文献:
    [42] 邹博伟,钱忠,陈站成,朱巧明,周国栋.面向自然语言文本的否定性与不确定性信息抽取.软件学报,2016,27(2):309-328. http://www.jos.org.cn/1000-9825/4860.htm[doi:10.13328/j.cnki.jos.004860]
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

钱忠,李培峰,周国栋,朱巧明.基于双向LSTM网络的不确定和否定作用范围识别.软件学报,2018,29(8):2427-2447

复制
分享
文章指标
  • 点击次数:3584
  • 下载次数: 5498
  • HTML阅读次数: 1942
  • 引用次数: 0
历史
  • 收稿日期:2017-03-18
  • 最后修改日期:2017-08-27
  • 录用日期:2017-11-08
  • 在线发布日期: 2018-01-09
文章二维码
友情链接:中国科学院软件研究所中国计算机学会中国科学院国家自然科学基金委员会CCF数字图书馆Int'l J of Softw Info计算机系统应用
您是第19785593位访问者
版权所有：中国科学院软件研究所 京ICP备05046678号-3
地址：北京市海淀区中关村南四街4号,邮政编码：100190
电话：010-62562563 传真：010-62562533 Email：jos@iscas.ac.cn
技术支持：北京勤云科技发展有限公司

京公网安备 11040202500063号