摘要
近年来电影行业蓬勃发展,相关的信息抽取和分析技术日益受到行业内的重视,其中对电影主创人物的分析尤为重要。而电影评论作为观影群体的主要反馈信息,具有重要的分析价值。如何从影评中自动抽取主创人名成为重要的基础工作。然而评论中观众对人物的称谓方式多样复杂,而且新电影的影评中往往存在大量人名未登录词,传统方法难以有效识别。针对影评的这些特点,该文提出一种基于多特征Bi-LSTM-CRF的影评人名识别方法。该方法通过利用外部人名语料和未标注影评提取字符级的特征,并采用Bi-LSTM-CRF模型进行人名字符序列标注。实验结果表明,该方法能够有效识别影评中的复杂称谓和人名未登录词,从而有效地抽取影评中的人名实体。
Person name in the movie reviews is featured by abbreviations and neologisms,which decreases the performances of classical models(e.g.CRF).To deal with this issue,this paper proposes a novel person name recognition method named Multi-Feature Bi-LSTM-CRF Model.This model extracts relevant character-level features by using external corpora and unlabeled reviews,then applies Bi-LSTM-CRF to identify the sequence of person names.The experimental results show that our model can effectively identify different forms of person names in the movie reviews.
引文
[1]Sundheim B M.Name Entity task definition,verson2.1[C]//Proceedings of the 6th Message Understanding Conference,1995:319-332.
[2]刘鸿宇,等.评价对象抽取及其倾向性分析[J].中文信息学报,2010,24(01):84-88.
[3]李中国,刘颖.边界模板和局部统计相结合的中国人名识别[J].中文信息学报,2006,20(05):44-50.
[4]倪吉,等.基于可信度模型的中文人名识别研究[J].中文信息学报,2011,25(03):45-50.
[5]曹波,苏一丹,邓琦.基于最大熵模型的中国人名自动识别[J].计算机工程与应用,2009,45(04):227-228.
[6]张素香,高国洋,戚银城.基于条件随机场的中国人名识别方法[J].郑州大学学报(理学版),2009,41(02):40-43.
[7]Chiu J P C,Nichols E.Named entity recognition with bidirectional LSTM-CNNs[J].Transactions of the Association for Computational Linguistics,2016,vol.4:357-370.
[8]Habibi M,et al.Deep learning with word embeddings improves biomedical named entity recognition[J].Bioinformatics,2017,33(14):i37-48.
[9]Pham T H,Le-Hong P.End-to-end recurrent neural network models for Vietnamese named entity recognition:Word-level vs.Character-level[J].Computational Linguistics,2017:219-232.
[10]Dong C,et al.Character-based LSTM-CRF with radical-level features for Chinese named entity recognition[M]//Natural Language Understanding and Intelligent Applications:Springer International Publishing,2016:239-250.
[11]Zhiheng Huang,Wei Xu,Kai Yu.Bidirectional LSTM-CRF models for sequence tagging.arXiv preprint arXiv:1508.01991,2015.
[12]Levy O,Goldberg Y.Dependency-based word embeddings[C]//Proceedings of Meeting of the Association for Computational Linguistics,2010:302-308.
[13]Chen X,et al.Joint learning of character and word embeddings[C]//Proceedings of International Conference on Artificial Intelligence.AAAI Press.2015:1236-1242.
[14]Tang D,et al.Learning sentiment-specific word embedding for Twitter sentiment classification[C]//Proceedings of Meeting of the Association for Computational Linguistics,2014:1555-1565.
[15]Mikolov T,et al.Distributed representations of words and phrases and their compositionality[J].Advances in Neural Information Processing Systems,2013,26:3111-3119.
[16]Lafferty J D,Mccallum A,Pereira F C N.Conditional random fields:Probabilistic models for segmenting and labeling sequence data[C]//Proceedings of the 18th International Conference on Machine Learning.Morgan Kaufmann Publishers.Inc.2001:282-289.
[17]王祖兴,吕钊,顾君忠.基于混合方法的中文人名识别研究[J].计算机工程与应用,2015,51(08):211-217.
[18]Pennington J,Socher R,Manning C.Glove:Global vectors for word representation[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2014:1532-1543.
[19]Kingma D P,Ba J L.Adam:a method for stochastic optimization[C]//Proceedings of the International Conference on Learning Representations.2015:1-15.
[20]Srivastava N,et al.Dropout:A simple way to prevent neural networks from overfitting[J].Journal of Machine Learning Research,2014,15(1):1929-1958.
[21]Dey R,Salemt F M.Gate-variants of gated recurrent unit(GRU)neural networks[C]//Proceedings of the International Midwest Symposium on Circuits and Systems.IEEE,2017:1597-1600.
(1)http://movie.weibo.com