Cross-Language Fake News Detection

Abstract With increasing globalization, news from different countries, and even in languages, has become readily available a way for many people to learn about other cultures. As around the world more reliant on social media, impact of fake public society also increases. However, most detection research focuses only English. In this work, we compared difference between textual features languages (Chinese English) their effect detecting news. We explored cross-language transmissibility models. found that Chinese are complex with English features. Our results illustrated bidirectional encoder representations transformers (BERT) model outperformed algorithms within-language data sets. sets, our findings demonstrated monitoring across is potentially feasible, while models trained inclusive language would perform better detection.