News-ExplaiNER: An Automated Fake News Detector and Explainer with Deep Learning and NLP

The widespread of the fake news on online platforms has emerged as a major concern in today’s information era. To counteract the spread of misinformation, researchers have put in a lot of efforts in researching effective and accurate fake news detection models. However, the lack of transparency and explainability hinders those models’ usage. This paper presents a novel methodology on detecting the veracity of the news contents and simultaneously providing an explanation on the prediction result, thereby enhancing transparency and trustworthiness of the model. This work proposes the News-ExplaiNER framework, an automated fake news detection and explanation system. The framework comprises of a news content classification model and a result explanation model. The classification achieved an accuracy score of $\mathbf{8 9. 4 4 \%}$, surpassing previous endeavours in this domain by 4%. Additionally, our explanation model implements a mechanism to quantify the performance of explainability in this particular task.