Tuned bidirectional encoder representations from transformers for fake news detection

Online medias are currently the dominant source of Information due to not being limited by time and place, fast wide distributions. However, inaccurate news, or often referred as fake news is a major problem in dissemination for online medias. Inaccurate information that true, engineered cover real has no factual basis. Usually, made form mass appeal presented guise genuine legitimate nuances deceive change reader's mind opinion. Identification from can be done with natural language processing (NLP) technologies. In this paper, we proposed bidirectional encoder representations transformers (BERT) identification. BERT model based on deep learning technologies it found effective many NLP tasks. study, use transfer fine-tuning adapt The experiments show our method could achieve accuracy 99.23%, recall 99.46%, precision 98.86%, F-Score 99.15%. It largely better than traditional same