Automatic identification of sarcasm in tweets and customer reviews
Journal of Intelligent & Fuzzy SystemsPublished 22 November 2019
Farah Naz, Muhammad Kamran, Waqar Mehmood, Wilayat Khan, Mohammed Saeed Alkatheiri, Ahmed Alghamdi
Citations12
SJR quartileQ2
SJR score0.36
SNIP0.62
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A mechanism for detecting sarcasm automatically using a classifier is proposed and the results of this study were quite promising and proved the effectiveness of the approach.
Abstract
The figurative language involving sarcasm on social networks is evolving the way how the humans use computers to communicate. Consequently, artificial intelligence techniques are applied in various scenarios to make the social networking more intelligent - for instance, identification of figurative language. Identifying both literal and non-literal meaning is not easy for a machine and it is hard even for people. Therefore, novel and exact frameworks ready to identify figurative languages are important. In sarcasm detection, this is even more challenging because sarcasm changes the polarity of an evidently positive or negative expression into its inverse. To maintain a distance for a sarcastic message being comprehended in its unintended actual meaning, in micro-blogging sites, for example messages on Twitter, sarcasm is frequently set apart with a hashtag for example, ’#sarcastic’, '#sarcasm', ’#not’ etc. Moreover, the customer reviews may also contain some element of sarcasm. To contribute to this area, we gathered the data of tweets and reviews from Twitter, thesarcasmdetector.com, and Kaggle and proposed a mechanism for detecting sarcasm automatically using a classifier. A detailed experimental study was also conducted to evaluate the proposed mechanism. The results of this study were quite promising and proved the effectiveness of our approach.
Keywords
Computer Science
Proceedings of the AAAI Conference on Artificial IntelligenceRecurrent Convolutional Neural Networks for Text Classification
2,306 Citations2015Siwei Lai, Liheng Xu +2 more
A recurrent convolutional neural network is introduced for text classification without human-designed features to capture contextual information as far as possible when learning word representations, which may introduce considerably less noise compared to traditional window-based neural networks.
Sarcasm as Contrast between a Positive Sentiment and Negative Situation
591 Citations2013Ellen Riloff, Ashequl Qadir +4 more
This work develops a sarcasm recognizer that automatically learns lists of positive sentiment phrases and negative situation phrases from sarcastic tweets and shows that identifying contrasting contexts using the phrases learned through bootstrapping yields improved recall for sarcasm recognition.
ACM Computing SurveysAutomatic Sarcasm Detection
393 Citations2017Aditya Joshi, Pushpak Bhattacharyya +1 more
Sarcasm Detection on Twitter
364 Citations2015Ashwin Rajadesingan, Reza Zafarani +1 more
This paper employs theories from behavioral and psychological studies to construct a behavioral modeling framework tuned for detecting sarcasm on Twitter and identifies behavioral traits intrinsic to users expressing sarcasm using the user's past tweets.
Proceedings of the International AAAI Conference on Web and Social MediaContextualized Sarcasm Detection on Twitter
356 Citations2021David Bamman, Noah Smith
By including extra-linguistic information from the context of an utterance on Twitter — such as properties of the author, the audience and the immediate communicative environment — this work is able to achieve gains in accuracy compared to purely linguistic features in the detection of this complex phenomenon, while also shedding light on features of interpersonal interaction that enable sarcasm in conversation.
Who cares about Sarcastic Tweets? Investigating the Impact of Sarcasm on Sentiment Analysis.
347 Citations2014Diana Maynard, Mark Greenwood
An analysis of the effect of sarcasm scope on the polarity of tweets, and a number of rules which enable the accuracy of sentiment analysis when sarcasm is known to be present are compiled.
Harnessing Context Incongruity for Sarcasm Detection
325 Citations2015Aditya Joshi, Vinita Sharma +1 more
A computational system that harnesses context incongruity as a basis for sarcasm detection is presented and it is shown how the features can capture intersentential incongrouity.
Fracking Sarcasm using Neural Network
293 Citations2016Aniruddha Ghosh, Tony Veale
A neural network semantic model composed of Convolution Neural Network (CNN) and followed by a Long short term memory (LSTM) network and finally a Deep neural network(DNN) is proposed, yielding an F-score of .92.
Proceedings of the American Society for Information Science and TechnologyA new perspective on Twitter hashtag use: Diffusion of innovation theory
266 Citations2010Hsia‐Ching Chang
Diffusion of innovation (DoI) is a theory that helps to explain the adoption process of an innovation by modeling its entire life cycle according to the aspects of communications and human information interactions and offers valuable insights into interface design that supports Twitter hashtag use and access.
arXiv (Cornell University)Semi-supervised Convolutional Neural Networks for Text Categorization via Region Embedding
249 Citations2015Rie Johnson, Tong Zhang
Modelling Sarcasm in Twitter, a Novel Approach
179 Citations2014Francesco Barbieri, Horacio Saggion +1 more
A novel computational model capable to detect sarcasm in the social network Twitter (a popular microblogging service which allows users to post short messages), which is easy to implement and, unlike previous systems, does not include patterns of words as features.
UWE Research Repository (UWE Bristol)Sarcasm Detection on Czech and English Twitter
175 Citations2014Tomáš Ptáċek, Ivan Habernal +1 more
This work is the first attempt at sarcasm detection in the Czech language by creating a large Czech Twitter corpus consisting of 7,000 manually-labeled tweets and evaluating two classifiers with various combinations of features on both the Czech and English datasets.
International Conference on Computational LinguisticsWord Embeddings and Convolutional Neural Network for Arabic Sentiment Classification
155 Citations2016Abdelghani Dahou, Shengwu Xiong +3 more
A scheme of Arabic sentiment classification, which evaluates and detects the sentiment polarity from Arabic reviews and Arabic social media, is studied and it is shown that the proposed scheme outperforms the existed methods on 4 out of 5 balanced and unbalanced datasets.
International Conference on Computational LinguisticsTweet Sarcasm Detection Using Deep Neural Network
154 Citations2016Meishan Zhang, Yue Zhang +1 more
This work investigates the use of neural network for tweet sarcasm detection, and compares the effects of the continuous automatic features with discrete manual features.
PubMedSemi-supervised Convolutional Neural Networks for Text Categorization via Region Embedding.
145 Citations2015Rie Johnson, Tong Zhang
The proposed scheme for embedding learning is based on the idea of two-view semi-supervised learning, which is intended to be useful for the task of interest even though the training is done on unlabeled data.
Indonesian social media sentiment analysis with sarcasm detection
127 Citations2013Edwin Lunando, Ayu Purwarianti
This work proposed two additional features to detect sarcasm after a common sentiment analysis is conducted, the features are the negativity information and the number of interjection words.
Knowledge-Based SystemsExtracting relevant knowledge for the detection of sarcasm and nastiness in the social web
109 Citations2014Raquel Justo, Thomas Corcoran +3 more
The results show that the sarcasm detection task benefits from the inclusion of linguistic and semantic information sources, while nasty language is more easily detected using only a set of surface patterns or indicators.
Text Classification with Topic-based Word Embedding and Convolutional Neural Networks
51 Citations2016Haotian Xu, Ming Dong +4 more
This paper proposes a novel neural language model, Topic-based Skip-gram, to learn topic-based word embeddings for biomedical literature indexing with CNNs, and describes two multimodal CNN architectures, which are able to employ different kinds of wordembeddings at the same time for text classification.
Polarity detection of sarcastic political tweets
37 Citations2014Devendra K. Tayal, Sumit Yadav +3 more
Two algorithms are presented, one to identify a sarcastic tweet and other to perform polarity detection on political sarcastic tweets, which are used to know the dissatisfaction of people against a particular party, candidate or government.
Recognition of Sarcasms in Tweets Based on Concept Level Sentiment Analysis and Supervised Learning Approaches
32 Citations2014Piyoros Tungthamthiti, Shirai Kiyoaki +1 more
A new method to identify sarcasm in tweets that focuses on several approaches: 1) sentiment analysis, 2) concept level and common-sense knowledge 3) coherence and 4) machine learning classification.
Lecture notes in social networksSarcasm Analysis on Twitter Data Using Machine Learning Approaches
32 Citations2017Santosh Kumar Bharti, Ramkrushna Pradhan +2 more
In this chapter, four approaches were proposed, namely parsing-based lexical generation algorithm, likes and dislikes contradiction, tweet contradicting universal facts, and tweets contradicting temporary facts, to extract text features such as lexical, hyperbole, behavioral, and universal facts.
Procedia Computer ScienceComparative Study of Different Sarcasm Detection Algorithms Based On Behavioral Approach
22 Citations2018Ravinder Ahuja, Shantanu Bansal +3 more
12 classification algorithms are applied on 4 types of datasets and gradient boosting is found to give the best accuracy in all 3 cases of a split ratio-50:50, 25:75, 10:90, and 85.03%).
Studies in fuzziness and soft computingSarcasm Detection Approaches for English Language
12 Citations2019Pragya Katyayan, Nisheeth Joshi
This chapter aims to be an introduction to the world of sarcasm and the methods of detecting it, and gives the reader a complete sense of the role of sarcasms in the field of sentiment analysis and how machines can be made capable of understanding sarcasm.
Cognitive systems monographsSarcasm Detection Using Contextual Incongruity
5 Citations2018Aditya Joshi, Pushpak Bhattacharyya +1 more
This chapter uses sequence labeling as a novel formulation of sarcasm detection to capture contextual incongruity in the dialogue of a monologue where multiple participants take part in a conversation.