Sarcasm as Contrast between a Positive Sentiment and Negative Situation

Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 704–714, Seattle, Washington, USA, 18-21 October 2013. c⃝2013 Association for Computational Linguistics Sarcasm as Contrast between a Positive Sentiment and Negative Situation Ellen Riloff, Ashequl Qadir, Prafulla Surve, Lalindra De Silva, Nathan Gilbert, Ruihong Huang School Of Computing University of Utah Salt Lake City, UT 84112 {riloff,asheq,alnds,ngilbert,huangrh}@cs.utah.edu, prafulla.surve@gmail.com Abstract A common form of sarcasm on Twitter con- sists of a positive sentiment contrasted with a negative situation. For example, many sarcas- tic tweets include a positive sentiment, such as “love” or “enjoy”, followed by an expression that describes an undesirable activity or state (e.g., “taking exams” or “being ignored”). We have developed a sarcasm recognizer to iden- tify this type of sarcasm in tweets. We present a novel bootstrapping algorithm that automati- cally learns lists of positive sentiment phrases and negative situation phrases from sarcastic tweets. We show that identifying contrast- ing contexts using the phrases learned through bootstrapping yields improved recall for sar- casm recognition. 1 Introduction Sarcasm is generally characterized as ironic or satir- ical wit that is intended to insult, mock, or amuse. Sarcasm can be manifested in many different ways, but recognizing sarcasm is important for natural lan- guage processing to avoid misinterpreting sarcastic statements as literal. For example, sentiment anal- ysis can be easily misled by the presence of words that have a strong polarity but are used sarcastically, which means that the opposite polarity was intended. Consider the following tweet on Twitter, which in- cludes the words “yay” and “thrilled” but actually expresses a negative sentiment: “yay! it’s a holi- day weekend and i’m on call for work! couldn’t be more thrilled! #sarcasm.” In this case, the hashtag #sarcasm reveals the intended sarcasm, but we don’t always have the benefit of an explicit sarcasm label. In the realm of Twitter, we observed that many sarcastic tweets have a common structure that creates a positive/negative contrast between a senti- ment and a situation. Specifically, sarcastic tweets often express a positive sentiment in reference to a negative activity or state. For example, consider the tweets below, where the positive sentiment terms are underlined and the negative activity/state terms are italicized. (a) Oh how I love being ignored. #sarcasm (b) Thoroughly enjoyed shoveling the driveway today! :) #sarcasm (c) Absolutely adore it when my bus is late #sarcasm (d) I’m so pleased mom woke me up with vacuuming my room this morning. :) #sarcasm The sarcasm in these tweets arises from the jux- taposition of a positive sentiment word (e.g., love, enjoyed, adore, pleased) with a negative activity or state (e.g., being ignored, bus is late, shoveling, and being woken up). The goal of our research is to identify sarcasm that arises from the contrast between a positive sen- timent referring to a negative situation. A key chal- lenge is to automatically recognize the stereotypi- cally negative “situations”, which are activities and states that most people consider to be unenjoyable or undesirable. For example, …

Content truncated to 500 words