Evaluation of Accuracy Degradation Resulting from Concept Drift in a Fake News Detection System Using Emotional Expression

Fake news on social media has become a problem. refers to false information that is deliberately intended deceive people. Several studies have been conducted automatic detection systems reduce the damage caused by fake news. However, most address improvements made in accuracy, and real-world operations are rarely discussed. As contents expressions of change over time, model with high accuracy loses after few years. This phenomenon called concept drift. conventional methods employ word representations, these exhibit degradation resulting from changes fads usage. using sentiment words can identify inflammatory sentences, which characteristic news, may suppress performance In this study, vector representations obtained an emotion dictionary was compared embedding. Subsequently, we verified resistance degradation. The results revealed method representation less susceptible Models learning achieve both enable further development systems.