GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media

By Yi-Ju Lu, Cheng-Te Li. Annual Meeting of the Association for Computational Linguistics, 2020

464 citations recorded by the source corpus.

This paper solves the fake news detection problem under a more realistic scenario on social media. Given the source short-text tweet and the corresponding sequence of retweet users without text comments, we aim at predicting whether the source tweet is fake or not, and generating explanation by highlighting the evidences on suspicious retweeters and the words they concern. We develop a novel neural network-based model, Graph-aware Co-Attention Networks (GCAN), to achieve the goal. Extensive experiments conducted on real tweet datasets exhibit that GCAN can significantly outperform state-of-the-art methods by 16% in accuracy on average. In addition, the case studies also show that GCAN can produce reasonable explanations.

Subfields: Artificial Intelligence, Sociology and Political Science, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/7b2ab7a828a6ae5cadffe79b1b3aa8bfbe3ae577

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Sociology and Political Science

In Statistical and Nonlinear Physics

Citations in the corpus

Explainable Automated Fact-Checking: A Survey (2020, methodology): “Lu and Li (2020) take a slightly different approach to generating explanations.”
User Preference-aware Fake News Detection (2021, background): “The list of ten profile feature names can be found in [8, 15] We also add two baselines that apply MLP directly on news textual embeddings encoded by word2vec and BERT.”
A General Survey on Attention Mechanisms in Deep Learning (2022, background): “combination with graph networks for the purpose of, for example, reading comprehension across multiple documents [41] and fake news detection [42].”
Formalizing Trust in Artificial Intelligence: Prerequisites, Causes and Goals of Human Trust in AI (2020, background): “dividuals Fairness checklists Bias exposure [65] Assess the impact on society and democracy Fairness checklists Explanations designed for applications such as fact checking [3] or fake news detection [47] Accountability Auditability of algorithms/data/design Factsheets (lineage) - Minimize and repor”
Suicide Ideation Detection via Social and Temporal User Representations using Hyperbolic Learning (2021, background): “While graph neural networks (GNNs) have made advances in enhancing NLP models for various tasks (Mishra et al., 2019a; Del Tredici et al., 2019; Lu and Li, 2020), two broad shortcomings limit their effectiveness for suicide ideation detection.”
A Comprehensive Survey on Graph Anomaly Detection With Deep Learning (2021, methodology): “…relationships between different types of nodes in heterogeneous graphs for anomaly detection, representative works, such as HGATRD [157], GCAN [156] and GLAN [158], typically decompose a heterogeneous graph into individual graphs according to meta-paths, e.g., one with tweets and users, and…”