Comparison of Deep Learning Models for Automatic Detection of Sarcasm Context on the MUStARD Dataset

Sentiment analysis is a major area of natural language processing (NLP) research, and its sub-area sarcasm detection has received growing interest in the past decade. Many approaches have been proposed, from basic machine learning to multi-modal deep solutions, progress made. Context proven be instrumental for many techniques that use context identify emerged. However, no NLP research focused on sarcasm-context as main topic. Therefore, this paper proposes an approach automatic context, aiming develop models can correctly contexts which may occur or appropriate. Using established dataset, MUStARD, multiple are trained benchmarked find best performer detection. This attention-based long short-term memory architecture achieves F1 score 60.1. Furthermore, we tested performance model SARC dataset compared it with other results reported literature better assess effectiveness approach. Future directions study opened, prospect developing conversational agent could even respond sarcasm.