Comparative Study of Different Sarcasm Detection Algorithms Based On Behavioral Approach
Ravinder Ahuja, Shantanu K. Bansal, Shuvam Prakash, Karthik Venkataraman, Alisha Banga · Procedia Computer Science · 2018
Sarcasm changes the dichotomy of an apparently negative or positive utterance into its contrary. While a fair amount of work has been done on automatically detecting emotion in human speech, there has been little research on sarcasm detection. In this paper, we have applied 12 classification algorithms (Gradient Boosting, Gaussian Naive Bayes, Adaboost etc.) on 4 types of datasets (Set1, Set2, Set3, Set4) and varied the split ratio of the datasets to check for the accuracy of every algorithm in different situations. We have applied the behavioral approach to sarcasm detection on twitter dataset. In set 4 were, we found gradient boosting to give the best accuracy in all 3 cases of a split ratio-50:50, 25:75, 10:90 (85.14%, 85.71%, and 85.03%).