Unsupervised Fake News Detection on Social Media: A Generative Approach

  title={Unsupervised Fake News Detection on Social Media: A Generative Approach},
  author={Shuo Yang and Kai Shu and Suhang Wang and Renjie Gu and Fan Wu and Huan Liu},
  booktitle={AAAI Conference on Artificial Intelligence},
Social media has become one of the main channels for people to access and consume news, due to the rapidness and low cost of news dissemination on it. [] Key Method We treat truths of news and users’ credibility as latent random variables, and exploit users’ engagements on social media to identify their opinions towards the authenticity of news. We leverage a Bayesian network model to capture the conditional dependencies among the truths of news, the users’ opinions, and the users’ credibility.

Figures and Tables from this paper

Beyond News Contents: The Role of Social Context for Fake News Detection

A tri-relationship embedding framework TriFN is proposed, which models publisher-news relations and user-news interactions simultaneously forfake news classification and significantly outperforms other baseline methods for fake news detection.

A Review of Fake News Detection Methods using Machine Learning

This survey deals with a review of existing machine learning algorithms Naïve Bayes, Convolutional Neural Network, LSTM, Neural network, Support Vector Machine proposed for detecting and reducing fake news from different social media platforms like Facebook, whatsapp, twitter, etc.

Evaluating the effectiveness of publishers’ features in fake news detection on social media

An algorithm, namely CreditRank, is proposed for evaluating publishers’ credibility on social networks and a high accurate multi-modal framework, namely FR-Detect, is suggested for fake news detection using user-related and content-related features.

Unsupervised Fake News Detection: A Graph-based Approach

GTUT, a graph-based approach for unsupervised fake news detection, is developed which operates in three phases and establishes the improved effectiveness of the method over state-of-the-art techniques for the task.

Incorporating User-Comment Graph for Fake News Detection

A novel detection framework, namely graph comment-user advanced learning framework (GCAL) is proposed in this paper, which outperforms 8 state-of-the-art baseline methods for fake news detection and is also explainable.

Hierarchical Propagation Networks for Fake News Detection: Investigation and Exploitation

This work builds hierarchical propagation networks for fake news and true news pieces, and performs a comparative analysis of the propagation network features from structural, temporal, and linguistic perspectives between fake and real news to demonstrate the potential of utilizing these features to detect fake news.

FakeNewsNet: A Data Repository with News Content, Social Context and Dynamic Information for Studying Fake News on Social Media

To facilitate fake news related researches, a fake news data repository FakeNewsNet is provided, which contains two comprehensive datasets that includes news content, social context, and spatiotemporal information.

Survey on Fake news and Fake profiles detection techniques

Four techniques are reviewed and compared in terms of accuracy for detection of fake news and profiles and among these four algorithms, Logistic regression performs better according to the review.


This article proposes a novel deep neural network that can detect fake news with greater than 90% accuracy within 5 minutes after it starts to spread and before it is retweeted 50 times, which is significantly faster than state-of-the-art baselines.



Fake News Detection on Social Media: A Data Mining Perspective

This survey presents a comprehensive review of detecting fake news on social media, including fake news characterizations on psychology and social theories, existing algorithms from a data mining perspective, evaluation metrics and representative datasets, and future research directions for fake news detection on socialMedia.

Exploiting Tri-Relationship for Fake News Detection

This paper explores the correlations of publisher bias, news stance, and relevant user engagements simultaneously, and proposes a Tri-Relationship Fake News detection framework (TriFN), and provides two comprehensive real-world fake news datasets to facilitate fake news research.

Tracing Fake-News Footprints: Characterizing Social Media Messages by How They Propagate

A novel approach to modeling the propagation of messages in a social network, TraceMiner, to infer embeddings of social media users with social network structures and utilize an LSTM-RNN to represent and classify propagation pathways of a message.

CSI: A Hybrid Deep Model for Fake News Detection

This work proposes a model called CSI which is composed of three modules: Capture, Score, and Integrate, and incorporates the behavior of both parties, users and articles, and the group behavior of users who propagate fake news.

News Verification by Exploiting Conflicting Social Viewpoints in Microblogs

This paper discovers conflicting viewpoints in news tweets with a topic model method, and builds a credibility propagation network of tweets linked with supporting or opposing relations that generates the final evaluation result for news.

News Credibility Evaluation on Microblog with a Hierarchical Propagation Model

This work proposes a hierarchical propagation model for evaluating news credibility on Micro blog by formulating this propagation process as a graph optimization problem, and provides a globally optimal solution with an iterative algorithm.

“Liar, Liar Pants on Fire”: A New Benchmark Dataset for Fake News Detection

This paper presents LIAR: a new, publicly available dataset for fake news detection, and designs a novel, hybrid convolutional neural network to integrate meta-data with text to improve a text-only deep learning model.

Information credibility on twitter

There are measurable differences in the way messages propagate, that can be used to classify them automatically as credible or not credible, with precision and recall in the range of 70% to 80%.

Leveraging the Crowd to Detect and Reduce the Spread of Fake News and Misinformation

A flexible representation of the above procedure using the framework of marked temporal point processes is introduced and a scalable online algorithm, CURB, is developed to select which stories to send for fact checking and when to do so to efficiently reduce the spread of misinformation with provable guarantees.

Faking Sandy: characterizing and identifying fake images on Twitter during Hurricane Sandy

The role of Twitter, during Hurricane Sandy (2012) to spread fake images about the disaster was highlighted, and automated techniques can be used in identifying real images from fake images posted on Twitter.