PL-NCC: a novel approach for fake news detection through data augmentation

Due to the rise of social media, the spread of fake news has been a serious threat to information integrity and the credibility of news sources. To address this challenge, there is an urgent need to develop robust and effective strategies to curb the spread of false information. There are two key contributions to fake news research in this work. First, we introduce the Psycho-Linguistic News Content and Comments (PL-NCC) dataset, a consolidated dataset derived from two prominent fake news datasets, NELA-GT and Fakeddit. Our dataset leverages linguistic and psychological features from both news articles and user comments to enhance the classification accuracy of benchmark models. Second, we propose the News Content and Comments (NCC) classification model, which utilizes the psychological features extracted from our PL-NCC dataset. By incorporating a feed-forward layer into a deep learning model, this approach enhances the effectiveness of the extracted features for more accurate fake news detection. Our model achieves a classification accuracy of over 90%, surpassing several existing baseline results.