FANG: Leveraging Social Context for Fake News Detection Using Graph Representation

TL;DR

FANG introduces a scalable graph-based social context model for fake news detection that outperforms existing methods and generalizes well to related tasks, especially with limited training data.

Contribution

FANG is a novel scalable graph representation learning framework that effectively captures social context for fake news detection, improving performance and robustness.

Findings

FANG achieves higher accuracy than recent models.

FANG is robust with limited training data.

FANG's representations generalize to other reporting factuality tasks.

Abstract

We propose Factual News Graph (FANG), a novel graphical social context representation and learning framework for fake news detection. Unlike previous contextual models that have targeted performance, our focus is on representation learning. Compared to transductive models, FANG is scalable in training as it does not have to maintain all nodes, and it is efficient at inference time, without the need to re-process the entire graph. Our experimental results show that FANG is better at capturing the social context into a high fidelity representation, compared to recent graphical and non-graphical models. In particular, FANG yields significant improvements for the task of fake news detection, and it is robust in the case of limited training data. We further demonstrate that the representations learned by FANG generalize to related tasks, such as predicting the factuality of reporting of a news medium.