Prof. Pradeep Kumar Roy’s research paper entitled ‘An Advanced Learning Approach for Detecting Sarcasm in Social Media Posts: Theory and Solutions’ has been published in the Wiley Social Science Quarterly journal.
The users of social media platforms view and share their daily life events through text, photographs, or videos. Such platforms receive many sarcastic posts daily because there are fewer limits on what can be posted. The presence of multiple languages and media types in a single post makes it difficult to identify the sarcastic messages on the current platform than on posts written solely in English. This research provides a theory and solutions for sarcastic post detection on social platforms. Hindi–English code-mixed data used to train and test the automated models for sarcasm detection makes this study unique in the context of traditional Machine Learning, Deep Neural Networks, LSTM (long short-term memory), CNN (convolutional neural network), and the combinations of BERT (Bidirectional Encoder Representations from Transformers) with LSTM. The experimental results confirm that the CNN, LSTM, and BERT-LSTM ensemble perform best for sarcasm detection in the Hindi-English code-mixed dataset. The proposed model achieved an accuracy of 96.29% and outperformed by 2.29% compared to the existing models.