情感分析是自然语言处理领域的重要研究问题。现有方法往往难以克服样本偏置与领域依赖问题,严重制约了情感分析的发展和应用。为此,该文提出了一种基于深度表示学习和高斯过程知识迁移学习的情感分析方法。该方法首先利用深度神经网络获得文本样本的分布式表示,而后基于深度高斯过程,从辅助数据中迁移与测试集数据分布相符的高质量样例扩充训练数据集用于分类器训练,以此提高文本情感分类系统性能。在COAE2014文本情感分类数据集上进行的实验结果显示,该文提出的方法可以有效提高文本情感分类性能,同时可以有效缓解训练数据的样本偏置以及领域依赖问题的影响。
Abstract
Sentiment analysis is an important topic in natural language processing research. Most existing sentiment analysis techniques are difficult to handle the domain dependent and sample bias issues, which restrain the development and application of sentiment analysis. To address these issues, this paper presents a sentiment analysis approach based on deep representation learning and Gaussian Processes transfer learning. Firstly, the distributed representations of text samples are learned based on deep neural network. Next, based on deep Gaussian processes, this approach selects quality samples with the distribution similar to testing dataset from additional dataset to expand the training dataset. The sentiment classifier trained on the expanded dataset is expected to achieve higher performance. The experimental results on COAE2014 dataset show that the proposed approach improved the sentiment classification performance. Meanwhile, this approach alleviates the influences of training sample bias and domain dependence.
关键词
情感分析 /
深度表示学习 /
高斯过程 /
迁移学习
{{custom_keyword}} /
Key words
sentiment analysis /
deep representation learning /
Gaussian processes /
transfer learning
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Xu R F, Wong K F, Xia Y. Coarse-Fine Opinion Mining-WIA in NTCIR-7 MOAT Task[C]//Proceedings of NTCIR 2008: 307-313.
[2] Kim Y. Convolutional Neural Networks for Sentence Classification[C]//Proceedings of EMNLP 2014: 1746-1751.
[3] Wang S, Manning C D. Baselines and Bigrams: Simple, Good Sentiment and Topic Classification[C]//Proceedings of ACL 2012: 90-94.
[4] Taboada M, Brooke J, Tofiloski M, et al. Lexicon-based Methods for Sentiment Analysis[J]. Computational Linguistics, 2011, 37(2): 267-307.
[5] Fu G, Wang X. Chinese Sentence-level Sentiment Classification Based on Fuzzy Sets[C]//Proceedings of ACL 2010: 312-319.
[6] Bollegala D, Weir D, Carroll J. Cross-domain Sentiment Classification Using a Sentiment Sensitive Thesaurus[J]. IEEE TKDE, 2013, 25(8): 1719-1731.
[7] Pang B, Lee L, Vaithyanathan S. Thumbs up?: Sentiment Classification Using Machine Learning Techniques[C]//Proceedings of ACL 2002: 79-86.
[8] Zhao J, Liu K, Wang G. Adding Redundant Features for CRFs-based Sentence Sentiment Classification[C]//Proceedings of ACL 2008: 117-126.
[9] Socher R, Perelygin A, Wu J Y., et al. Recursive Deep Models for Semantic Compositionality over a Sentiment Treebank[C]//Proceedings of EMNLP 2013: 1631-1642.
[10] Xiao S, Chengcheng L, Fuji R. Sentiment Analysis for Chinese Microblog based on Deep Neural Networks with Convolutional Extension Features[J]. Neurocomputing, 2016, 210: 227-236.
[11] Pan S J, Yang Q. A Survey on Transfer Learning[J]. IEEE TKDE, 2010, 22(10): 1345-1359.
[12] Dai W, Yang Q, Xue G. R, et al. Boosting for Transfer Learning[C]//Proceedings of ICML 2007: 193-200.
[13] Argyriou A, Evgeniou T, Pontil M. Convex Multi-task Feature Learning[C]//Proceedings of NIPS 2007: 19-41
[14] Zadrozny B. Learning and Evaluating Classifiers under Sample Selection Bias[C]//Proceedings of ICML 2004: 114-121.
[15] Fan W, Davidson I, Zadrozny B, et al. An Improved Categorization of Classifiers Sensitivity on Sample Selection Bias[C]//Proceedings of ICDM 2005: 605-608
[16] Blitzer J, McDonald R, Pereira F. Domain Adaptation with Structural Correspondence Learning[C]//Proceedings of ACL 2006: 120-128.
[17] Xu R F, Xu J, Wang X. Instance Level Transfer Learning for Cross Lingual Opinion Analysis[C]//Proceedings of ACL 2011: 182-188.
[18] Gui L, Xu R.F, Lu Q, et al. Cross-lingual Opinion Analysis via Negative Transfer Detection[C]//Proceedings of ACL 2014: 860-865.
[19] Gui L, Lu Q, Xu R.F, et al. A Novel Class Noise Estimation Method and Application in Classification[C]//Proceedings of CIKM 2015: 1081-1090.
[20] Kandemir M. Asymmetric Transfer Learning with Deep Gaussian Processes[C]//Proceedings of ICML 2015: 730-738.
[21] Snelson E., Ghahramani Z. Sparse Gaussian Processes Using Pseudo-inputs[C]//Proceedings of NIPS 2005: 1257-1264.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61370165);国家863计划(2015AA015405);深圳市孔雀计划技术创新项目(KQCX20140521144507925);深圳市基础研究项目(JCYJ20150625142543470);广东省数据科学工程技术研究中心开放课题(2016KF09)
{{custom_fund}}
PDF(2142 KB)
吴冬茵(1992—),硕士研究生,主要研究领域为自然语言处理、文本情绪计算。E-mail: wudongyinhit@gmail.com
桂林(1988—),博士研究生,主要研究领域为机器学习、情感分析。E-mail: guilin.nlp@gmail.com
陈钊(1990—),硕士,助理工程师,主要研究领域为复杂网络、信息传播、自然语言处理、文本情绪计算。E-mail: gilbertchen@126.com
