An empirical perspective on sarcasm detection in tweets using machine learning techniques

The majority of the time, Sentiment Analysis is utilised to figure out what the author is thinking about. Major provocation has been occurring, and irony detection is believed to be one of the most provocative aspects of the situation currently. Irony is a unique manner of describing information that is in opposition to the topic being discussed, resulting in ambiguity. Data preparation, which covers numerous techniques such as lemmatization, tokenization, and stemming, is a main activity performed by the majority of software engineers. Many studies are carried out in the field of irony detection, which covers a variety of feature extraction algorithms. Support Vector Machine (SVM), linear regression, Nave Bayes, Random Forest, and other machine learning classifierswere employed in these studies, among others. The accuracy, precision, recall, and F-score obtained as a result of these research efforts may be utilized to forecast the most appropriate model for a given situation. In this study, we will explore the numerous methodologies that are employed in ironic text identification for Sentiment Analysis.