基于深度学习的多元文本情感研究与分析
Research and Analysis of Textual Multi-Emotion Based on Deep Learning
DOI: 10.12677/CSA.2018.85076, PDF,  被引量   
作者: 陈 楠*, 陈进才, 卢 萍：华中科技大学武汉光电国家研究中心，湖北 武汉
关键词: 情感分析；文本分类；多元情感；情感词向量；Emotion Analysis； Text Classification； Multiple Emotion； Emotional Word Vector
摘要: 文本情感分析主要是通过文本挖掘技术对带有倾向性的文本进行情感分析和处理，识别其中主观性文本的倾向是正面、负面、中性的过程，这种关于文本情感颗粒的划分是不充分的，不全面的，显得过于生硬和暴力，不仅不能有效地体现出不同的文本情感颗粒的强度和大小，而且还需要大量的人工标注。本文针对此问题，提出和构建了基于Co-Training半监督训练的多元文本情感数据集，并且结合情感词频、情感词典、情感语义信息构建了D & W、T & W、SSW三种情感词向量，最后利用CNN和LSTM神经网络结构模型分别对构建的多元数据集进行了情感词向量的对比训练和模型优化，从而验证了情感词向量的有效性，而且提升了文本情感分类的准确度。
Abstract: Text emotional analysis is mainly based on text mining technology for emotional analysis and processing of the text with tendentiousness, which is the subjective text tendency recognition process of positive and negative, neutral. This text about emotional particle division is not sufficient, not comprehensive, is too stiff and violence, and not only cannot effectively reflect the text sentiment granules with different strength and size, but also needs a lot of manual annotation. This paper proposes and constructs the multiple text sentiment data Co-Training based on semi supervised training set, and combines with the emotion of frequency, emotion dictionary, emotion semantic in-formation to construct the D & W, T & W, SSW three kinds of emotion word vector. Finally, CNN and LSTM neural network structure model is used to construct multivariate data sets that were com-pared with the training and optimization model of emotion word vector, which verifies the validity of the emotional word vector, but also improves the accuracy of text sentiment classification.
文章引用：陈楠, 陈进才, 卢萍. 基于深度学习的多元文本情感研究与分析[J]. 计算机科学与应用, 2018, 8(5): 669-686. https://doi.org/10.12677/CSA.2018.85076

参考文献

[1] Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986) Learning Internal Representations by Error Propagation. MIT Press, Nature, 3l8-362.
[2] Bengio, Y., Ducharme, R., Vincent, P., et al. (2003) A Neural Probabilistic Language Model. Journal of Machine Learning Research, 3, 1137-1155.
[3] Collobert, R. and Weston, J. (2008) A Unified Ar-chitecture for Natural Language Processing: Deep Neural Networks with Multitask Learning. Proceedings of the Twen-ty-Fifth International Conference on Machine Learning (ICML 2008), Helsinki, Finland, 160-167.
[4] Hinton, G.E. (1986) Learning Distributed Representations of Concepts. Proceedings of the Eighth Annual Conference of the Cognitive Science Society, 1, 12.
[5] Mikolov, T., Sutskever, T., et al. (2013) Distributed Representations of Words and Phrases and Their Compositionality. Advances in Neural Information Processing Systems, 26, 3111-3119.
[6] 梁军, 柴玉梅, 原慧斌, 等. 基于深度学习的微博情感分析[J]. 中文信息学报, 2014, 28(5): 155-161.
[7] 陈翠平. 基于深度信念网络的文本分类算法[J]. 计算机系统应用, 2015, 24(2): 121-126.
[8] Kim, Y. (2014) Convolutional Neural Net-works for Sentence Classification. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, 25-29 October 2014, 1746-1751. [Google Scholar] [CrossRef] 
[9] Pang, B., Lee, L. and Vaithyanathan, S. (2002) Sentiment Classification Using Machine Learning Techniques. Proceedings of the Conference on Empirical Methods in Natural Language Pro-cessing (EMNILP), July 2002, Philadelphia, 79-86.
[10] Davidov, D., Tsur, O. and Rappoport, A. (2010) Enhanced Sentiment Learning Using Twitter Hashtags and Smileys. International Conference on Computational Linguistics: Post-ers, Paris, 241-249.
[11] 李婷婷, 姬东鸿. 基于SVM和CRF多特征组合的微博情感分析[J]. 计算机应用研究, 2015, 32(4): 978-981.
[12] 李荣陆, 王建会, 陈晓云, 等. 使用最大熵模型进行中文文本分类[J]. 计算机研究与发展, 2005, 42(1): 94-101.
[13] Taboada, M., Brooke, J., Tofiloski, M., et al. (201l) Lexicon-Based Methods for Sen-timent Analysis. Computational Linguistics, 37, 267-307.
[14] Hu, M. and Liu, B. (2004) Mining and Summarizing Customer Reviews. Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Seattle, Washington DC, USA, August, 168-177.