基于雙向LSTM的維吾爾語事件因果關系抽取

田生偉; 周興發(fā); 禹龍; 馮冠軍; 艾山?吾買爾; 李圃

doi:10.11999/JEIT170402

基于雙向LSTM的維吾爾語事件因果關系抽取

doi: 10.11999/JEIT170402 cstr: 32379.14.JEIT170402

1.
(新疆大學軟件學院烏魯木齊 830046) ②(新疆大學網絡中心烏魯木齊 830046) ③(新疆大學人文學院烏魯木齊 830046) ④(新疆大學信息科學與工程學院烏魯木齊 830046) ⑤(新疆大學語言學院烏魯木齊 830046)

基金項目:

國家自然科學基金(61662074, 61563051, 61262064)，國家自然科學基金重點項目(61331011)，新疆自治區(qū)科技人才培養(yǎng)項目(QN2016YX0051)

計量
- 文章訪問數: 2297
- HTML全文瀏覽量: 336
- PDF下載量: 258
- 被引次數: 0
出版歷程
- 收稿日期: 2017-05-02
- 修回日期: 2017-07-19
- 刊出日期: 2018-01-19

Causal Relation Extraction of Uyghur Events Based on Bidirectional Long Short-term Memory Model

1.
(School of Software, Xinjiang University, Urumqi 830046, China)
2.
(Net Center, Xinjiang University, Urumqi 830046, China)

Funds:

The National Natural Science Foundation of China (61662074, 61563051, 61262064), The Key Project of National Natural Science Foundation of China (61331011), Xinjiang Uygur Autonomous Region Scientific and Technological Personnel Training Project (QN2016YX0051)

摘要

摘要: 針對傳統(tǒng)方法不能有效抽取維吾爾語事件因果關系的問題，該文提出一種基于雙向LSTM(Bidirectional Long Short-Term Memory, BiLSTM)的維吾爾語事件因果關系抽取方法。通過對維吾爾語語言以及事件因果關系特點的研究，提取出10項基于事件內部結構信息的特征；同時為充分利用事件語義信息，引入詞嵌入作為BiLSTM的輸入，提取事件句隱含的深層語義特征并利用批樣規(guī)范化(Batch Normalization, BN)算法加速BiLSTM的收斂；最后融合這兩類特征作為softmax分類器的輸入進而完成維吾爾語事件因果關系抽取。實驗結果表明，該方法用于維吾爾語事件因果關系的抽取準確率為 89.19%, 召回率為 83.19%, F值為86.09%，證明了該文提出的方法在維吾爾語事件因果關系抽取上的有效性。
- 語言信號處理 /
- 事件因果關系 /
- 維吾爾語 /
- 雙向LSTM /
- 詞嵌入 /
- 批樣規(guī)范化
Abstract: Since the traditional events causal relation has the disadvantages of small recognition coverage, a method for causal relation extraction of Uyghur events is presented based on Bidirectional Long Short-Term Memory (BiLSTM) model. In order to make full use of the event structure information, 10 characteristics of the Uyghur events structure information are extracted based on the study of the events causal relationship and Uyghur language features; At the same time, the word embedding is introduced as the input of BiLSTM to extract the deep semantic features of the Uyghur events and Batch Normalization (BN) algorithm is usded to accelerate the convergence of BiLSTM. Finally, concatenating these two kinds of features as the input of the softmax classifier to extract the Uyghur events causal relations. This method is used in the causal relation extraction of Uyghur events, and the results show that the precision rate, the recall rate and F value can reach 89.19 %, 83.19% and 86.09 %, indicating the effectiveness and practicability of the method of causal relation extraction of Uyghur events.
- Language signal processing /
- Event causal relation /
- Uyghur /
- Bidirectional Long Short-Term Memory (BiLSTM) /
- Word embedding /
- Batch Normalization (BN)

HTML全文

參考文獻(22)

SHA L, LI S J, CHANG B B, et al. Joint learning teplates and slots for event schema induction[C]. Proceedings of the North American Chapter of the Association for Computational Linguistics-Human Language Technologies, California, USA, 2016: 428-434. doi: 10.18653/v1/N16-1049.

CHEN C and NG V. Joint modeling for chinese event extraction with rich linguistic features[C]. Proceedings of COLING 2012, Mumbai, India, 2012: 529-544.

CHEN Y, XU L, LIU K, et al. Event extraction via dynamic multi-pooling convolutional neural networks[C]. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Beijing, China, 2015: 167-176. doi: 10.3115/v1/P15-1017.

ZHANG Y J, LIU Z T, and ZHOU W, Event recognition based on deep learning in Chinese texts[J]. Plos One, 2016, 11(8): e0160147. doi: 10.1371/journal.pone.0160147.

CHANG C Y, TENG Z Y and ZHANG Y. Expectation- regulated neural model for event mention extraction[C]. Proceedings of NAACL-HLT, California, USA, 2016: 400-410. doi: 10.18653/v1/N16-1045.

干紅華, 潘云鶴. 一種基于事件的因果關系的結構分析方法[J]. 模式識別與人工智能, 2003, 16(1): 56-62. doi: 10. 3969/j.issn.1003-6059.2003.01.011.

Gan Honghua and Pan Yunhe. A new analysis of the structure of event causation[J]. Pattern Recognition and Artiflcial Intelligence, 2003, 16(1): 56-62. doi: 10.3969/j.issn. 1003-6059.2003.01.011.

GIRJU R and MOLDOVAN D. Text mining for causal relations[C]. Proceedings of the 15th International Florida Artiflcial Intelligence Research Society Conference, Florida, USA, 2002: 360-364.

MARCU D and ECHIHABI A. An unsupervised approach to recognizing discourse relations[C]. Proceedings of the 40th Annual Meeting on Association for Computational Linguistics, Philadelphia, USA, 2002: 368-375. doi: 10.3115/ 1073083.1073145.

SORGENTE A, VETTIGLI G, and MELE F. Automatic extraction of cause-effect relations in natural language text[C]. Proceedings of the 13th Conference of the Italian Association for Artificial Intellgence, Rome, Italian, 2013: 37-48.

付劍鋒, 劉宗田, 劉煒, 等. 基于層疊條件隨機場的事件因果關系抽取[J]. 模式識別與人工智能, 2011, 24(4): 567-573. doi: 10.3969/j.issn.1003-6059.2011.04.016.

FU Jianfeng, LIU Zongtian, LIU Wei, et al. Event causal relation extraction based on cascaded conditional random flelds[J]. Pattern Recognition and Artiflcial Intelligence, 2011, 24(4): 567-573. doi: 10.3969/j.issn.1003-6059.2011.04.016.

鐘軍, 禹龍, 田生偉, 等. 基于雙層模型的維語突發(fā)事件因果關系抽取[J]. 自動化學報, 2014, 40(4): 771-779. doi: 10.3724/ SP.J.1004.2013.00771.

Zhong Jun, Yu Long, Tian Shengwei, et al. Causal relation extraction of uyghur emergency events based on cascaded model[J]. Acta Automatica Sinica, 2014, 40(4): 771-779. doi: 10.3724/SP.J.1004.2013.00771.

TANG D, QIN B, FENG X, et al. Effective LSTMs for target-dependent sentiment classification[C]. Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, Osaka, Japan, 2016: 3298-3307.

ZHOU P, SHI W, TIAN J, et al. Attention-based bidirectional long short-term memory networks for relation classification[C]. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Berlin, Germany, 2016: 207-212. doi: 10.18653/v1/P162034.

賀宇, 潘達, 付國宏. 基于自動編碼特征的漢語解釋性意見句識別[J]. 北京大學學報(自然科學版), 2015, 51(2): 235240. doi: 10.13209/J.0479-8023.2015.041.

HE Yu, PAN Da and FU Guohong. Chinese explanatory opinionated sentence recognition based on autoEncoding features[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2015, 51(2): 235-240. doi: 10.13209/J.0479-8023. 2015.041.

MIKOLOV T, SUTSKEVER I, CHEN K, et al. Distributed representations of words and phrases and their compositionality[C]. Proceedings of Advances in Neural Information Processing Systems, Vancouve, Canada, 2013: 3111-3119.

FENG X C, HUANG L F, TANG D Y, et al. A language-independent neural network for event detection[C]. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Berlin, Germany, 2016: 66-71. doi: 10.18653/v1/P16-2011.

SUTSKEVER I, VINYALS O, and LE Q V. Sequence to sequence learning with neural networks[C]. Proceedings of Advances in Neural Information Processing Systems, Quebec, Canada, 2014: 3104-3112.

IOFFE S and SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]. Proceedings of the 32th International Conference on Machine Learning, Lille, France, 2015: 448-456.

施引文獻

資源附件(0)

訪問統(tǒng)計