基于联合特征与Bi-LSTM模型的英汉双语文本情绪预测
|
摘要
英汉双语文本中的情绪可以通过英语和汉语的单语或者双语形式来表达。然而,以往的研究主要集中在单语文本的情绪分析,只有少数研究侧重于英汉双语文本。为提高英汉双语文本情绪预测效果,本文结合情绪词典方法与深度学习方法,使用联合特征与Bi-LSTM模型来对英汉双语文本进行情绪预测。首先基于情绪词典抽取出双语文本中包含的情绪词特征,然后联合情绪词特征与双语文本特征输入至Bi-LSTM模型进行特征学习,最后将学习到的深度语义特征输入到分类器中进行情绪预测。实验结果表明,该方法对英汉双语文本的情绪预测有良好的效果。
Emotions in English-Chinese bilingual texts can be expressed in either monolingual or bilingual form in English and Chinese. However, the preceding researches have focused on analyzing emotions in monolingual text, and only a few of them have paid attention to English-Chinese bilingual texts. In order to improve the emotion prediction in English-Chinese bilingual texts, we combine the emotional dictionary and deep learning method using the Bi-LSTM joint feature model. Firstly, the emotional word feature is extracted from the bilingual text according to the emotional dictionary. And then, the deep semantic features are learned based on Bi-LSTM model with the joint features of the bilingual text and emotional words. Finally, the deep semantic features are fed to the classifier to make emotion predictions. The experimental results show that our method has made good performance on the emotion prediction of English-Chinese bilingual texts.
Emotions in English-Chinese bilingual texts can be expressed in either monolingual or bilingual form in English and Chinese. However, the preceding researches have focused on analyzing emotions in monolingual text, and only a few of them have paid attention to English-Chinese bilingual texts. In order to improve the emotion prediction in English-Chinese bilingual texts, we combine the emotional dictionary and deep learning method using the Bi-LSTM joint feature model. Firstly, the emotional word feature is extracted from the bilingual text according to the emotional dictionary. And then, the deep semantic features are learned based on Bi-LSTM model with the joint features of the bilingual text and emotional words. Finally, the deep semantic features are fed to the classifier to make emotion predictions. The experimental results show that our method has made good performance on the emotion prediction of English-Chinese bilingual texts.
引文
[1]DAS D,BANDYOPADHYAY S.Emotion Analysis on Social Media:Natural Language Processing Approaches and Applications[M].Heidelberg:Springer,2014:19-37.
[2]李寿山,李逸薇,刘欢欢,等.基于情绪相关事件上下文的隐含情绪分类方法研究[J].中文信息学报,2013,27(6):90-96.DOI:10.3969/j.issn.1003-0077.2013.06.013.LI S S,LI Y W,LIU H H,et al.Implicit emotion classification with the context of emotion related event[J].Journal of Chinese Information,2013,27(6):90-96.DOI:10.3969/j.issn.1003-0077.2013.06.013(Ch).
[3]LEE S,WANG Z.Emotion in code-switching texts:Corpus construction and analysis[C]//Proceedings of the 8th Sighan Workshop on Chinese Language Processing.Beijing:Association for Computational Linguistics,2015:91-99.DOI:10.18653/v1/W15-3116.
[4]WANG Z,LEE S Y M,LI S,et al.Emotion analysis in code-switching text with joint factor graph model[J].IEEE/ACM Transactions on Audio Speech&Language Processing,2017,25(3):469-480.DOI:10.1109/TAS-LP.2016.2637280.
[5]MA C,OSHERENKO A,PRENDINGER H,et al.Achat system based on emotion estimation from text and embodied conversational messengers[C]//Proceedings of the4th International Conference on Entertainment Computing.Heidelberg:Springer,2005:535-538.
[6]AMAN S,SZPAKOWICZ S.Identifying expressions of emotion in text[C]//Proceedings of the 10th International Conference on Text,Speech and Dialogue.Heidelberg:Springer,2007:196-205.DOI:10.1007/978-3-540-74628-7_27.
[7]XU J,XU R F,ZHENG Y Z,et al.Chinese emotion lexicon developing via multi-lingual lexical resources integration[C]//Computational Linguistics and Intelligent Text Processing.Heidelberg:Springer,2013:174-182.DOI:10.1007/978-3-642-37256-8_15.
[8]WU F Z,HUANG Y F,SONG Y Q,et al.Towards building a high-quality microblog-specific Chinese sentiment lexicon[J].Decision Support Systems,2016,87:39-49.DOI:10.1016/j.dss.2016.04.007.
[9]WEN S Y,WAN X J.Emotion classification in microblog texts using class sequential rules[C]//Proceedings of the28th AAAI Conference on Artificial Intelligence.Québec City:AAAI Press,2014:187-193.
[10]READ J.Using emoticons to reduce dependency in machine learning techniques for sentiment classification[C]//Proceedings of the ACL Student Research Workshop.Pennsylvania:Association for Computational Linguistics,2005:43-48.DOI:10.3115/1628960.1628969.
[11]OUYANG X,ZHOU P,LI C H,et al.Sentiment analysis using convolutional neural network[C]//Proceedings of the 13th IEEE International Conference on Pervasive Intelligence and Computing.Liverpool:IEEE,2015:2359-2364.DOI:10.1109/CIT/IUCC/DASC/PICO-M.2015.349.
[12]SANTOS C,GATTI M.Deep convolutional neural networks for sentiment analysis of short texts[C]//Proceedings of the 25th International Conference on Computational Linguistics.New York:ACM,2014:69-78.
[2]李寿山,李逸薇,刘欢欢,等.基于情绪相关事件上下文的隐含情绪分类方法研究[J].中文信息学报,2013,27(6):90-96.DOI:10.3969/j.issn.1003-0077.2013.06.013.LI S S,LI Y W,LIU H H,et al.Implicit emotion classification with the context of emotion related event[J].Journal of Chinese Information,2013,27(6):90-96.DOI:10.3969/j.issn.1003-0077.2013.06.013(Ch).
[3]LEE S,WANG Z.Emotion in code-switching texts:Corpus construction and analysis[C]//Proceedings of the 8th Sighan Workshop on Chinese Language Processing.Beijing:Association for Computational Linguistics,2015:91-99.DOI:10.18653/v1/W15-3116.
[4]WANG Z,LEE S Y M,LI S,et al.Emotion analysis in code-switching text with joint factor graph model[J].IEEE/ACM Transactions on Audio Speech&Language Processing,2017,25(3):469-480.DOI:10.1109/TAS-LP.2016.2637280.
[5]MA C,OSHERENKO A,PRENDINGER H,et al.Achat system based on emotion estimation from text and embodied conversational messengers[C]//Proceedings of the4th International Conference on Entertainment Computing.Heidelberg:Springer,2005:535-538.
[6]AMAN S,SZPAKOWICZ S.Identifying expressions of emotion in text[C]//Proceedings of the 10th International Conference on Text,Speech and Dialogue.Heidelberg:Springer,2007:196-205.DOI:10.1007/978-3-540-74628-7_27.
[7]XU J,XU R F,ZHENG Y Z,et al.Chinese emotion lexicon developing via multi-lingual lexical resources integration[C]//Computational Linguistics and Intelligent Text Processing.Heidelberg:Springer,2013:174-182.DOI:10.1007/978-3-642-37256-8_15.
[8]WU F Z,HUANG Y F,SONG Y Q,et al.Towards building a high-quality microblog-specific Chinese sentiment lexicon[J].Decision Support Systems,2016,87:39-49.DOI:10.1016/j.dss.2016.04.007.
[9]WEN S Y,WAN X J.Emotion classification in microblog texts using class sequential rules[C]//Proceedings of the28th AAAI Conference on Artificial Intelligence.Québec City:AAAI Press,2014:187-193.
[10]READ J.Using emoticons to reduce dependency in machine learning techniques for sentiment classification[C]//Proceedings of the ACL Student Research Workshop.Pennsylvania:Association for Computational Linguistics,2005:43-48.DOI:10.3115/1628960.1628969.
[11]OUYANG X,ZHOU P,LI C H,et al.Sentiment analysis using convolutional neural network[C]//Proceedings of the 13th IEEE International Conference on Pervasive Intelligence and Computing.Liverpool:IEEE,2015:2359-2364.DOI:10.1109/CIT/IUCC/DASC/PICO-M.2015.349.
[12]SANTOS C,GATTI M.Deep convolutional neural networks for sentiment analysis of short texts[C]//Proceedings of the 25th International Conference on Computational Linguistics.New York:ACM,2014:69-78.