fake-news

Rumor detection

• WWW2018 Detect Rumor and Stance Jointly by Neural Multi-task Learning

  • paper
  • determining the stances on the posts in question can be pertinent to the successful detection of rumors, and vice versa.
  • propose a joint framework that unifies the two highly pertinent tasks, i.e., rumor detection and stance classification.
  • dataset: rumor detection-Twitter (Liu et al.) stance classification-PHEME and FNC

• CIKM2018 Rumor Detection with Hierarchical Social Attention Network

  • paper
  • we detect rumors by leveraging hierarchical representations at different levels and the social contexts.
  • we propose a novel hierarchical neural network combined with social information (word-level, post-level, subevent-level)
  • dataset: Twitter and Weibo (Ma et al.)

• NLPCC2019 Rumor Detection with Hierarchical Recurrent Convolutional Neural Network

  • paper
  • Usually, the events on social media are divided into several time segments, and for each segment, corresponding text will be converted as vectors for various neural network models to detect rumors. During this process, however, only sentence-level embedding has been considered, while the contextual information at the word level has been largely ignored.
  • we propose a novel rumor detection method based on a hierarchical recurrent convolutional neural network, which integrates contextual information for rumor detection.
  • dataset: Twitter and Weibo (Ma et al.)

• ICDM2019 Rumor Stance Classification via Machine Learning with Text, User and Propagation Features

  • Stances are commonly divided into four categories: support, deny, query and comment
  • we conduct an in-depth study on the feature engineering for this task.
  • dataset: RumorEval

• AAAI2020 Capturing the Style of Fake News

  • paper
  • code
  • explore automatic methods that can detect online documents of low credibility, especially fake news, based on the style they are written in. Since fake news sources usually attempt to attract attention for a shortterm financial or political goal, they favour informal, sensational, affective language.
  • model: BiLSTMAvg and a model based on Stylometric
  • the credibility of news articles in our corpus can indeed be estimated based on the style they are written in.

• AAAI2020 Weak Supervision for Fake News Detection via Reinforcement Learning

  • paper
  • code

• CHI2020 Fake News on Facebook and Twitter: Investigating How People (Don't) Investigate