Identification of Information Calumny using Sarcasm Features

With extensive access to the internet, information gets spread rapidly, leading to a generation of misinformation affecting the life of people, businesses, etc. The researchers have been keen on identifying solutions to this problem of misinformation generation. Machine learning and deep learning-based approaches are becoming popular to develop automated models to identify information calumny. The machine learning-based models are based on a feature extraction process. Handcrafted features play a crucial role in model development. This includes content or text and user-specific features. One type of significant feature is sarcasm, which is used to communicate emotions in a way opposite to what it actually means to be. This feature can undoubtedly assist in identifying the credibility of contents and information calumny. Previous researchers have studied the sarcasm in fake news but have not used these as features to develop models. Authors have proposed a novel approach to identify information calumny with sarcasm to fill this gap. The proposed method showed the highest accuracy of 93.88%.