Application of BiMPM Technique to Detect the Fake News

The rise of fake news has had a worldwide impact, influencing both social behaviors and political settings. While many systems aimed at detecting fake news primarily rely on linguistic patterns, they often falter when dealing with content that’s complex and needs a grasp of the current context. The study presents a novel system from Korea for identifying fake news, which taps into a “Fact DB” —a collection of facts confirmed through human assessment. To cross-check statements, this system employs a deep learning technique named BiMPM. Yet, BiMPM struggles with extensive texts and unknown word associations. To mitigate these challenges, the study suggests a unique technique that melds article summarization with entity correlation, in tandem with BiMPM. The results highlight enhanced accuracy in identifying false news through this method.