Sarcasm as Contrast between a Positive Sentiment and Negative Situation

By Ellen Riloff, Ashequl Qadir, Prafulla Surve, Lalindra De Silva, Nathan Gilbert, Ruihong Huang. Conference on Empirical Methods in Natural Language Processing, 2013

696 citations recorded by the source corpus.

A common form of sarcasm on Twitter consists of a positive sentiment contrasted with a negative situation. For example, many sarcastic 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 identify this type of sarcasm in tweets. We present a novel bootstrapping algorithm that automatically learns lists of positive sentiment phrases and negative situation phrases from sarcastic tweets. We show that identifying contrasting contexts using the phrases learned through bootstrapping yields improved recall for sarcasm recognition.

Subfields: Artificial Intelligence, Experimental and Cognitive Psychology

View Paper: https://www.semanticscholar.org/paper/b0e31bf1d900a4d33c2c233d84e6275568f1e3d3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Experimental and Cognitive Psychology

Citations in the corpus

Contextualized Sarcasm Detection on Twitter (2015, background): “These purely text-based approaches can be surprisingly accurate across different domains (Carvalho et al. 2009; Davidov, Tsur, and Rappoport 2010; González-Ibáñez, Muresan, and Wacholder 2011; Riloff et al. 2013; Lukin and Walker 2013; Reyes, Rosso, and Veale 2013), but are divorced from any notion ”
Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Research (2020, neutral): “[182], and hashtags, Maynard & Greenwood [137].”
Revealing Persona Biases in Dialogue Systems (2021, neutral): “Also, we observe that VADER does not always properly account for negations and sarcasm—behaviors observed in existing sentiment literature (Reitan et al. 2015; Riloff et al. 2013).”
Implicitly Abusive Language – What does it actually look like and why are we not getting there? (2021, background): “Riloff et al. (2013) demonstrated that mining for a particular linguistic construction is an effective means to recognize a speciﬁc type of sarcasm.”
Humans Require Context to Infer Ironic Intent (so Computers Probably do, too) (2014, methodology): “Elsewhere, Riloff et al. (2013) proposed a method that exploits contrasting sentiment in the same utterance to detect irony.”
Harnessing Context Incongruity for Sarcasm Detection (2015, methodology): “Our approach is architecturally similar to Tsur et al. (2010b) who use a semi-supervised pattern acquisition followed by classification.”
Sarcasm Detection on Twitter: A Behavioral Modeling Approach (2015, background): “Current research on sarcasm detection on Twitter [38, 10, 20, 29] has primarily focused on obtaining information from the text of the tweets.”
Characterizing and Detecting Hateful Users on Twitter (2018, background): “However, in OSNs, the meaning of such content is often not self-contained, referring, for instance, to some event which just happened and the texts are packed with informal language, spelling errors, special characters and sarcasm (Dhingra et al. 2016; Riloff et al. 2013).”
Fracking Sarcasm using Neural Network (2016, methodology): “Riloff (Riloff et al., 2013) and Liebrecht (Liebrecht et al., 2013) applied N-grams features to a classiﬁer for English and Dutch tweets and observed that some topics recur frequently in sarcastic tweets, such as schools, dentists, church life, public transport, the weather and so on.”
Data Statements for Natural Language Processing: Toward Mitigating System Bias and Enabling Better Science (2018, background): “The latter includes annotations for sentiment (Liu, 2012) and for ﬁgurative language or sarcasm (e.g., Riloff et al., 2013; Ptáˇcek et al., 2014).”
SemEval-2018 Task 3: Irony Detection in English Tweets (2018, methodology): “Regardless of the methodology and algorithm used, irony detection often involves binary clas-siﬁcation where irony is deﬁned as instances that express the opposite of what is meant (e.g. Riloff et al., 2013; Joshi et al., 2017).”
R3: Reverse, Retrieve, and Rank for Sarcasm Generation with Commonsense Knowledge (2020, background): “While most computational work has focused on sarcasm detection (Davidov et al., 2010; González-Ibáñez et al., 2011; Riloff et al., 2013; Ghosh et al., 2015; Joshi et al., 2015b; Muresan et al., 2016; Ghosh and Veale, 2017; Ghosh et al., 2017, 2018), research on sarcasm generation is in its infancy (”
Multi-Modal Sarcasm Detection in Twitter with Hierarchical Fusion Model (2019, methodology): “Earlier methods extract carefully engineered discrete features from texts (Davidov et al., 2010; Riloff et al., 2013; Pt´aˇcek et al., 2014; Bouazizi and Ohtsuki, 2015), including n-grams, word’s sentiment, punctuations, emoticons, part-of-speech tags, etc.”
A Survey on Hate Speech Detection using Natural Language Processing (2017, background): “It remains to be seen, whether in the future new computational approaches can actually solve these problems or whether hate speech is a research problem similar to sarcasm where only certain subtypes have been shown to be automatically detected with the help of NLP (Riloff et al., 2013).”
Data Statements for NLP: Toward Mitigating System Bias and Enabling Better Science (2018, background): “The latter includes annotations for sentiment (Liu, 2012) and for figurative language or sarcasm (e.g. Riloff et al. 2013; Ptáček et al. 2014).”