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.
@article{arxiv.2012.09118,
title = {Exploring Thematic Coherence in Fake News},
author = {Martins Samuel Dogo and Deepak P and Anna Jurek-Loughrey},
journal= {arXiv preprint arXiv:2012.09118},
year = {2020}
}
Comments
10 pages, 1 figure, to be published in Proceedings of the 8th International Workshop on News Recommendation and Analytics (INRA 2020)