Towards Bootstrapping a Polarity Shifter Lexicon using Linguistic Features

We present a major step towards the creation of the first high-coverage lexicon of polarity shifters. In this work, we bootstrap a lexicon of verbs by exploiting various linguistic features. Polarity shifters, such as ‘abandon’, are similar to negations (e.g. ‘not’) in that they move the polarity of a phrase towards its inverse, as in ‘abandon all hope’. While there exist lists of negation words, creating comprehensive lists of polarity shifters is far more challenging due to their sheer number. On a sample of manually annotated verbs we examine a variety of linguistic features for this task. Then we build a supervised classifier to increase coverage. We show that this approach drastically reduces the annotation effort while ensuring a high-precision lexicon. We also show that our acquired knowledge of verbal polarity shifters improves phrase-level sentiment analysis.

[1]  Roser Morante,et al.  Descriptive Analysis of Negation Cues in Biomedical Texts , 2010, LREC.

[2]  Eneko Agirre,et al.  Personalizing PageRank for Word Sense Disambiguation , 2009, EACL.

[3]  Jiawei Han,et al.  Data-Driven Contextual Valence Shifter Quantification for Multi-Theme Sentiment Analysis , 2016, CIKM.

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

[5]  Noah A. Smith,et al.  Semi-Supervised Frame-Semantic Parsing for Unknown Predicates , 2011, ACL.

[6]  Claire Cardie,et al.  Identifying Expressions of Opinion in Context , 2007, IJCAI.

[7]  Vincent Ng,et al.  Frame Semantics for Stance Classification , 2013, CoNLL.

[8]  Maite Taboada,et al.  Lexicon-Based Methods for Sentiment Analysis , 2011, CL.

[9]  Michael Wiegand,et al.  Opinion Holder and Target Extraction based on the Induction of Verbal Categories , 2015, CoNLL.

[10]  Guodong Zhou,et al.  Tree Kernel-based Negation and Speculation Scope Detection with Structured Syntactic Parse Features , 2013, EMNLP.

[11]  Christopher Potts,et al.  Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank , 2013, EMNLP.

[12]  Hinrich Schütze,et al.  Classification of Inconsistent Sentiment Words using Syntactic Constructions , 2012, COLING.

[13]  Christopher D. Manning,et al.  Introduction to Information Retrieval , 2010, J. Assoc. Inf. Sci. Technol..

[14]  Yejin Choi,et al.  ConnotationWordNet: Learning Connotation over the Word+Sense Network , 2014, ACL.

[15]  Sanda M. Harabagiu,et al.  Negation, Contrast and Contradiction in Text Processing , 2006, AAAI.

[16]  L. Brinton.  VERB PARTICLES IN ENGLISH: ASPECT OR AKTIONSART?* , 1985 .

[17]  Saif M. Mohammad,et al.  The Effect of Negators, Modals, and Degree Adverbs on Sentiment Composition , 2016, WASSA@NAACL-HLT.

[18]  Yang Huang,et al.  A novel hybrid approach to automated negation detection in clinical radiology reports. , 2007, Journal of the American Medical Informatics Association : JAMIA.

[19]  Iryna Gurevych,et al.  Supersense Embeddings: A Unified Model for Supersense Interpretation, Prediction, and Utilization , 2016, ACL.

[20]  Margaret Mitchell,et al.  Overview of the TAC2013 Knowledge Base Population Evaluation: English Sentiment Slot Filling , 2013, TAC.

[21]  Minhwan Yu,et al.  Deep Semantic Frame-Based Deceptive Opinion Spam Analysis , 2015, CIKM.

[22]  Eduard Hovy,et al.  Extracting Opinions, Opinion Holders, and Topics Expressed in Online News Media Text , 2006 .

[23]  Rada Mihalcea,et al.  Integrating Knowledge for Subjectivity Sense Labeling , 2009, NAACL.

[24]  Claire Cardie,et al.  Learning with Compositional Semantics as Structural Inference for Subsentential Sentiment Analysis , 2008, EMNLP.

[25]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[26]  Janyce Wiebe,et al.  Lexical Acquisition for Opinion Inference: A Sense-Level Lexicon of Benefactive and Malefactive Events , 2014, WASSA@ACL.

[27]  Claire Cardie,et al.  Automatically Creating General-Purpose Opinion Summaries from Text , 2011, RANLP.

[28]  Janyce Wiebe,et al.  +/-EffectWordNet: Sense-level Lexicon Acquisition for Opinion Inference , 2014, EMNLP.

[29]  Bonnie L. Webber,et al.  Neural Networks For Negation Scope Detection , 2016, ACL.

[30]  Dietrich Klakow,et al.  A survey on the role of negation in sentiment analysis , 2010, NeSp-NLP@ACL.

[31]  Janyce Wiebe,et al.  Benefactive/Malefactive Event and Writer Attitude Annotation , 2013, ACL.

[32]  Partha Pratim Talukdar,et al.  Weakly-Supervised Acquisition of Labeled Class Instances using Graph Random Walks , 2008, EMNLP.

[33]  Andrea Esuli,et al.  Determining the semantic orientation of terms through gloss classification , 2005, CIKM '05.

[34]  Daisuke Ikeda,et al.  Learning to Shift the Polarity of Words for Sentiment Classification , 2008, IJCNLP.

[35]  John B. Lowe,et al.  The Berkeley FrameNet Project , 1998, ACL.

[36]  Massimo Poesio,et al.  Negation of protein-protein interactions: analysis and extraction , 2007, ISMB/ECCB.

[37]  Taher H. Haveliwala.  Topic-sensitive PageRank , 2002, IEEE Trans. Knowl. Data Eng..

[38]  János Csirik,et al.  The BioScope corpus: annotation for negation, uncertainty and their scope in biomedical texts , 2008, BioNLP.

[39]  Noah A. Smith,et al.  Probabilistic Frame-Semantic Parsing , 2010, NAACL.

[40]  J. R. Landis,et al.  The measurement of observer agreement for categorical data. , 1977, Biometrics.

[41]  Roser Morante,et al.  A Metalearning Approach to Processing the Scope of Negation , 2009, CoNLL.

[42]  Cristian Danescu-Niculescu-Mizil,et al.  Without a ’doubt’? Unsupervised Discovery of Downward-Entailing Operators , 2009, NAACL.

[43]  Thorsten Joachims,et al.  Making large scale SVM learning practical , 1998 .

[44]  M. Krifka.  Some Remarks on Polarity Items , 1991 .

[45]  Bing Liu,et al.  Opinion spam and analysis , 2008, WSDM '08.

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