Sarcastic or Not: Word Embeddings to Predict the Literal or Sarcastic Meaning of Words

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1003–1012, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Sarcastic or Not: Word Embeddings to Predict the Literal or Sarcastic Meaning of Words Debanjan Ghosh§ and Weiwei Guo† and Smaranda Muresan‡ §School of Communication and Information, Rutgers University, NJ, USA †Department of Computer Science, Columbia University, NY, USA ‡Center for Computational Learning Systems, Columbia University, NY, USA debanjan.ghosh@rutgers.edu, weiwei@cs.columbia.edu, smara@ccls.columbia.edu Abstract Sarcasm is generally characterized as a figure of speech that involves the substi- tution of a literal by a figurative mean- ing, which is usually the opposite of the original literal meaning. We re-frame the sarcasm detection task as a type of word sense disambiguation problem, where the sense of a word is either literal or sar- castic. We call this the Literal/Sarcastic Sense Disambiguation (LSSD) task. We address two issues: 1) how to collect a set of target words that can have either literal or sarcastic meanings depending on con- text; and 2) given an utterance and a target word, how to automatically detect whether the target word is used in the literal or the sarcastic sense. For the latter, we investi- gate several distributional semantics meth- ods and show that a Support Vector Ma- chines (SVM) classifier with a modified kernel using word embeddings achieves a 7-10% F1 improvement over a strong lex- ical baseline. 1 Introduction Recognizing sarcasm is important for understand- ing people’s actual sentiments and beliefs. For example, failing to recognize the following mes- sage as being sarcastic “I love that I have to go back to the emergency room”, will lead a senti- ment and opinion analysis system to infer that the author’s sentiment is positive towards the event of “going to the emergency room”. Current ap- proaches have framed the sarcasm detection task as predicting whether a full utterance is sarcastic or not (Davidov et al., 2010; Gonz´alez-Ib´a˜nez et al., 2011; Riloff et al., 2013; Liebrecht et al., 2013; Maynard and Greenwood, 2014). We propose a re-framing of sarcasm detection as a type of word sense disambiguation problem: given an utterance and a target word, identify whether the sense of the target word is literal or sarcastic. We call this the Literal/Sarcastic Sense Disambiguation (LSSD) task. In the above utter- ance, the word “love” is used in a sarcastic, non- literal sense (the author’s intended meaning be- ing most likely the opposite of the original literal meaning - a negative sentiment, such as “hate”). Two key challenges need to be addressed: 1) how to collect a set of target words that can have a lit- eral or a sarcastic sense, depending on context; and 2) given an utterance containing a target word, how can we determine whether the target word is used in its literal sense (e.g., “I love to take a nice stroll in the park every morning”), or in a sarcastic sense (e.g., “I love going to the dentist.”). To address the first challenge, we …

Content truncated to 500 words