评价词和评价对象抽取在意见挖掘中是一个重要的任务,我们在句子级评价词和评价对象联合抽取任务上研究了长短时记忆(long short-term memory)神经网络模型的几种变种应用。长短时记忆神经网络模型是一种循环神经网络模型,该模型使用长短时记忆模型单元作为循环神经网络的记忆单元,它能够获得更多的长距离上下文信息,同时避免了普通循环神经网络的梯度消失和梯度爆炸的问题。我们对比了传统的方法,实验结果证明长短时记忆神经网络模型优于以前的方法,在细粒度评价词和评价对象的联合抽取中达到更好的性能。
Abstract
To deal with opinion word and opinion target extraction, we explore several variants of long short-term memory recurrent neural networks for joint extraction of them at sentence-level. We also compare our models with previous classical approaches. The results of the experiments show that long short-term memory recurrent neural networks outperform previous baselines, achieving new state-of-the-art results for joint extraction of fine-grained opinion words and opinion targets.
关键词
循环神经网络 /
长短时记忆模型 /
评价词与评价对象联合抽取 /
深度学习 /
序列标注
{{custom_keyword}} /
Key words
recurrent neural networks /
long short-term memory model /
opinion words and target extraction /
deep learning /
sequence label
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Janyce Wiebe,Theresa Wilson, Claire Cardie.Annotating expressions of opinions and emotions in language[J].Language resources and evaluation,2005,[1] 39(2-3):165-210.
[2] Minqing Hu,Bing Liu.Mining and summarizing customer reviews[C]//Proceedings of the tenth SIGKDD,ACM,2004:168-177.
[3] Guang Qiu,Bing Liu,Jiajun Bu,et al.Opinion word expansion and target extraction through double propagation[J].Computational Linguistics,2011.37(1):9-27.
[4] Kang Liu,Liheng Xu,and Jun Zhao.Opinion target extraction using word-based translation model[C]//Proceedings of EMNLP.Association for Computational Linguistics.2012:1346-1356.
[5] Yejin Choi,Claire Cardie,Ellen Riloff,et al.Identifying sources of opinions with conditional random fields and extraction patterns[C]//Proceedings of EMNLP,2005:355-362.Association for Computational Linguistics.
[6] Ronan Collobert,Jason Weston.A unified architecture for natural language processing:Deep neural networks with multitask learning[C]//Proceedings of ICML,2008:160-167.ACM.
[7] Xinchi Chen,Xipeng Qiu,Chenxi Zhu,et al.Long short-term memory neural networks for Chinese word segmentation[C]//Proceedings of EMNLP.Association for Computational Linguistics,2015.
[8] Pengfei Liu,Shafiq Joty, Helen Meng.Finegrained opinion mining with recurrent neural networks and word embeddings[C]//Proceedings of EMNLP.Association for Computational Linguistics,2015.
[9] Sepp Hochreiter.The vanishing gradient problem during learning recurrent neural nets and problem solutions[J].International Journal of Uncertainty,Fuzziness and Knowledge-Based Systems,1998.6(02):107-116.
[10] Eric Breck,Yejin Choi, Claire Cardie.Identifying expressions of opinion in context[C]//IJCAI,volum 7,2007:2683-2688.
[11] Yejin Choi,Claire Cardie.Hierarchical sequential learning for extracting opinions and their attributes[C]//Proceedings of the ACL 2010 Conference Short Papers,2010:269-274.Association for Computational Linguistics.
[12] Richard Johansson,Alessandro Moschitti.Syntactic and semantic structure for opinion expression detection[C]//Proceedings of the Fourteenth Conference on Computational Natural Language Learning,2010:67-76.Association for Computational Linguistics.
[13] Richard Johansson,Alessandro Moschitti.Extracting opinion expressions and their polarities:exploration of pipelines and joint models[C]//Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics:Human Language Technologies:short papers-Volume 2,2011:101-106.Association for Computational Linguistics.
[14] Bishan Yang,Claire Cardie.Extracting opinion expressions with semi-markov conditional random fields[C]//Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning,2012:1335-1345.Association for Computational Linguistics.
[15] Bishan Yang,Claire Cardie.Joint infence for fine-grained opinion extraction[C]//ACL(1),2013:1640-1649.
[16] Ozan Irsoy,Claire Cardie.Opinion mining with deep recurrent neural networks[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing(EMNLP),2014:720-728.
[17] Tomas Mikolov,Stefan Kombrink,Lukas Burget,et al.Extensions of recurrent neural network language model[C]//Acoustics,Speech and Signal Processing(ICASSP),2011 IEEE International Conference on,2011:5528-5531.IEEE.
[18] Mike Schuster,Kuldip K Paliwal.Bidirectional recurrent neural networks.Signal Processing,IEEE Transactions on,1997,45(11):2673-2681.
[19] Zhiqiang Toh,Jian Su.Nlangp at semeval-2016 task 5:Improving aspect based sentiment analysis using neural network features[C]//Proceedings of SemEval,2016:282-288.
[20] Grégoire Mesnil,Xiaodong He,Li Deng,et al.Investigation of recurent-neural network architectures and learning methods for spoken language understanding[J].INTERSPEECH,2013:3771-3775.
[21] G David Forney Jr.The viterbi algorithm[C]//Proceedings of the IEEE,1973.61(3):268-278.
[22] Alex Graves,Schmidhuber.Framewise phoneme classification with bidirectional LSTM and other neural network architectures[J].Neural Networks,2005,18(5):602-610.
[23] Jingbo Zhu,Huizhen Wang,Benjamin K Tsou,et al.Multi-aspect opinion polling from textual reviews//Proceedings of the 18th ACM Conference on Information and Knowledge Management,2009:1799-1802.ACM.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61363032);海南省重大科技计划项目(ZDKJ2017012)
{{custom_fund}}
PDF(3932 KB)
