Exploring Thematic Coherence in Fake News
This research provides an incremental method for identifying fake news by analyzing thematic coherence, which could benefit fact-checkers and social media platforms.
This study analyzed the thematic coherence of fake news using topic models across seven cross-domain datasets. It found that fake news exhibits greater thematic deviation between its opening sentences and the rest of the article compared to truthful news.
The spread of fake news remains a serious global issue; understanding and curtailing it is paramount. One way of differentiating between deceptive and truthful stories is by analyzing their coherence. This study explores the use of topic models to analyze the coherence of cross-domain news shared online. Experimental results on seven cross-domain datasets demonstrate that fake news shows a greater thematic deviation between its opening sentences and its remainder.