Like Article, Like Audience: Enforcing Multimodal Correlations for Disinformation Detection

TL;DR

This paper presents a multimodal learning approach that leverages correlations between user-generated and shared content to improve disinformation detection in online news, without relying on user profiles during prediction.

Contribution

It introduces a novel multimodal algorithm that uses user content correlations during training to enhance fake news detection, avoiding user profiling at inference.

Findings

Improved accuracy in disinformation detection across multiple datasets.

Visualization shows better discrimination between fake and real news.

Model successfully applied to three neural classifiers.

Abstract

User-generated content (e.g., tweets and profile descriptions) and shared content between users (e.g., news articles) reflect a user's online identity. This paper investigates whether correlations between user-generated and user-shared content can be leveraged for detecting disinformation in online news articles. We develop a multimodal learning algorithm for disinformation detection. The latent representations of news articles and user-generated content allow that during training the model is guided by the profile of users who prefer content similar to the news article that is evaluated, and this effect is reinforced if that content is shared among different users. By only leveraging user information during model optimization, the model does not rely on user profiling when predicting an article's veracity. The algorithm is successfully applied to three widely used neural classifiers, and results are obtained on different datasets. Visualization techniques show that the proposed model learns feature representations of unseen news articles that better discriminate between fake and real news texts.