Sarcasm detection framework using context, emotion and sentiment features

Sarcasm detection is an essential task that can help identify the actual sentiment in user-generated data, such as discussion forums or tweets. a sophisticated form of linguistic expression because its surface meaning usually contradicts inner, deeper meaning. In this paper, we propose model, incorporates different features to capture sarcasm. We use pre-trained transformer and CNN context features, transformers on emotions analysis tasks. our architecture, amotion models were used only feature extractors. Other blocks (pre-trained CNN) fine-tuned. run experiments four datasets from domains. Our approach outperformed previous state-of-the-art results social networking platforms online media.