Sparsh Arora - Academia.edu (original) (raw)

Related Authors

NAFISA HOSSAIN NUJAT, 180041213

Computer Science & Information Technology (CS & IT) Computer Science Conference Proceedings (CSCP)

Ijariit Journal

AMISHA GANGWAR

Syful Islam

Mohamed LICHOURI

Mohamed LICHOURI

Université des Sciences et Technologie Houari Boumediene (USTHB)

Uploads

Papers by Sparsh Arora

Research paper thumbnail of Sarcasm Analysis Using Machine Learning

The growth of social media has been exponential within the recent years. Complete information is ... more The growth of social media has been exponential within the recent years. Complete information is being put out on the domains directly through social media. Sarcastic tweets can mislead processing activities and end in wrong classification. This paper compares various classification algorithms like Random Forest, SVC, Logistic Regression, Linear SVC and Gaussian Naïve Bayes to detect sarcasm within the input given by the user, the simplest classifier is chosen and paired with various pre-processing and filtering techniques using emojis to supply the simplest possible output. The emojis being the central idea introduced through this paper, the obtained results may be used as an input for other research and applications.

Research paper thumbnail of Sarcasm Analysis Using Machine Learning

The growth of social media has been exponential within the recent years. Complete information is ... more The growth of social media has been exponential within the recent years. Complete information is being put out on the domains directly through social media. Sarcastic tweets can mislead processing activities and end in wrong classification. This paper compares various classification algorithms like Random Forest, SVC, Logistic Regression, Linear SVC and Gaussian Naïve Bayes to detect sarcasm within the input given by the user, the simplest classifier is chosen and paired with various pre-processing and filtering techniques using emojis to supply the simplest possible output. The emojis being the central idea introduced through this paper, the obtained results may be used as an input for other research and applications.

Log In