Two-Path Deep Semisupervised Learning for Timely Fake News Detection

摘要: News in social media, such as Twitter, has been generated high volume and speed. However, very few of them are labeled (as fake or true news) by professionals near real time. In order to achieve timely detection news a novel framework two-path deep semisupervised learning (SSL) is proposed where one path for supervised the other unsupervised learning. The learns on limited amount data, while able learn huge unlabeled data. Furthermore, these two paths implemented with convolutional neural networks (CNNs) jointly optimized complete SSL. addition, we build shared CNN extract low-level features both data feed into paths. To verify this framework, implement Word CNN-based SSL model test it sets: LIAR PHEME. Experimental results demonstrate that built can recognize effectively