Variational Weakly Supervised Sentiment Analysis with Posterior Regularization

Sentiment analysis is an important task in natural language processing (NLP). Most of existing state-of-the-art methods are under the supervised learning paradigm. However, human annotations can be scarce. Thus, we should leverage more weak supervision for sentiment analysis. In this paper, we propose a posterior regularization framework for the variational approach to the weakly supervised sentiment analysis to better control the posterior distribution of the label assignment. The intuition behind the posterior regularization is that if extracted opinion words from two documents are semantically similar, the posterior distributions of two documents should be similar. Our experimental results show that the posterior regularization can improve the original variational approach to the weakly supervised sentiment analysis and the performance is more stable with smaller prediction variance.

[1]  Danqi Chen,et al.  A Fast and Accurate Dependency Parser using Neural Networks , 2014, EMNLP.

[2]  Janyce Wiebe,et al.  Recognizing Contextual Polarity in Phrase-Level Sentiment Analysis , 2005, HLT.

[3]  Chao Zhang,et al.  Weakly-Supervised Text Classification Using Label Names Only , 2020, EMNLP.

[4]  Bing Liu,et al.  Mining and summarizing customer reviews , 2004, KDD.

[5]  Christopher Potts,et al.  Learning Word Vectors for Sentiment Analysis , 2011, ACL.

[6]  Jiawei Han,et al.  Weakly-Supervised Neural Text Classification , 2018, CIKM.

[7]  Tao Li,et al.  A Non-negative Matrix Tri-factorization Approach to Sentiment Classification with Lexical Prior Knowledge , 2009, ACL.

[8]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[9]  Lei Zhang,et al.  Sentiment Analysis and Opinion Mining , 2017, Encyclopedia of Machine Learning and Data Mining.

[10]  Xiang Zhang,et al.  Character-level Convolutional Networks for Text Classification , 2015, NIPS.

[11]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[12]  Aitor García Pablos,et al.  V3: Unsupervised Aspect Based Sentiment Analysis for SemEval2015 Task 12 , 2015, *SEMEVAL.

[13]  Meena Nagarajan,et al.  Semantic Lexicon Expansion for Concept-Based Aspect-Aware Sentiment Analysis , 2014, SemWebEval@ESWC.

[14]  Ben Taskar,et al.  Posterior Regularization for Structured Latent Variable Models , 2010, J. Mach. Learn. Res..

[15]  TaboadaMaite,et al.  Lexicon-based methods for sentiment analysis , 2011 .

[16]  Peter D. Turney Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews , 2002, ACL.

[17]  Diego Marcheggiani,et al.  Discrete-State Variational Autoencoders for Joint Discovery and Factorization of Relations , 2016, TACL.

[18]  Philipp Koehn,et al.  Synthesis Lectures on Human Language Technologies , 2016 .

[19]  Mohand Boughanem,et al.  Using WordNet's Semantic Relations for Opinion Detection in Blogs , 2009, ECIR.

[20]  Jun Zhao,et al.  Sentiment Classification with Graph Co-Regularization , 2014, COLING.

[21]  Themis Palpanas,et al.  Scalable discovery of contradictions on the web , 2010, WWW '10.

[22]  Xin Liu,et al.  A Variational Approach to Weakly Supervised Document-Level Multi-Aspect Sentiment Classification , 2019, NAACL.

[23]  John Carroll,et al.  Weakly supervised techniques for domain-independent sentiment classification , 2009, TSA@CIKM.

[24]  M. de Rijke,et al.  UvA-DARE ( Digital Academic Repository ) Using WordNet to measure semantic orientations of adjectives , 2004 .

[25]  Iryna Gurevych,et al.  Using Distributional Similarity for Lexical Expansion in Knowledge-based Word Sense Disambiguation , 2012, COLING.