DUCS at SemEval-2022 Task 6: Exploring Emojis and Sentiments for Sarcasm Detection

Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022), pages 1005 - 1011 July 14-15, 2022 ©2022 Association for Computational Linguistics DUCS at SemEval-2022 Task 6: Exploring Emojis and Sentiments for Sarcasm Detection Vandita Grover Department of Computer Science University of Delhi vgrover@cs.du.ac.in Hema Banati Dyal Singh College University of Delhi hemabanati@dsc.du.ac.in Abstract This paper describes the participation of team DUCS at SemEval 2022 Task 6: iSarcasmEval - Intended Sarcasm Detection in English and Arabic. Team DUCS participated in SubTask A of iSarcasmEval which was to determine if the given English text was sarcastic or not. In this work, emojis were utilized to capture how they contributed to the sarcastic nature of a text. It is observed that emojis can augment or re- verse the polarity of a given statement. Thus sentiment polarities and intensities of emojis, as well as those of text, were computed to de- termine sarcasm. Use of capitalization, word repetition, and use of punctuation marks like ’!’ were factored in as sentiment intensifiers. An NLP augmenter was used to tackle the imbal- anced nature of the sarcasm dataset. Several architectures comprising of various ML and DL classifiers, and transformer models like BERT and Multimodal BERT were experimented with. It was observed that Multimodal BERT outper- formed other architectures tested and achieved an F1-score of 30.71%. The key takeaway of this study was that sarcastic texts are usually positive sentences. In general emojis with pos- itive polarity are used more than those with negative polarities in sarcastic texts. 1 Introduction According to the Collins Dictionary 1 "Sarcasm is speech or writing which actually means the opposite of what it seems to say". Fox Tree et al. (2020) report that sarcasm is chal- lenging to identify, even for humans. People often use gestures like rolling of eyes or heavy tonal stress to express sarcasm (Pandey et al., 2019), in speech or in-person communication. Intonation 1https://www.collinsdictionary.com/dictionary/english/sarcasm and stress in speech, too, are strong indicators of sarcasm (Castro et al., 2019). Also, the context in the form of shared knowledge between the speaker and the audience can be leveraged to detect sar- casm (Amir et al., 2016). But most of the time the tone or the context is missing in text data, es- pecially tweet data. Some users attempt to use capitalization like "OH YEAH" or repetitions like "woowww" to indicate tonal intensity. But, most of the time these cues are not enough, especially in absence of context to gauge sarcasm. This makes sarcasm detection a challenging task. In this work, team DUCS participated in SemEval 2022 Task 6 : iSarcasmEval Subtask A, (Abu Farha et al., 2022), to determine if the given text is sarcas- tic or not. The English text training dataset contains 3468 sentences. Each text is labeled as sarcastic or non-sarcastic by the text-author. Of the 3468 sen- tences 867 are sarcastic and 2601 are non-sarcastic. In the dataset corresponding to a sarcastic text, a rephrased non-sarcastic version has also been pro- vided. These sarcastic …

Content truncated to 500 words