FR-Detect: A Multi-Modal Framework for Early Fake News Detection on Social Media Using Publishers Features

TL;DR

FR-Detect is a multi-modal framework that leverages publisher features and content analysis to detect fake news early on social media with high accuracy.

Contribution

This work introduces novel publisher-related features and a multi-modal neural network framework for improved early fake news detection.

Findings

Publisher features improve detection accuracy by up to 13%.

The framework achieves up to 29% better F1-score.

Multi-modal approach enhances early fake news detection.

Abstract

In recent years, with the expansion of the Internet and attractive social media infrastructures, people prefer to follow the news through these media. Despite the many advantages of these media in the news field, the lack of any control and verification mechanism has led to the spread of fake news, as one of the most important threats to democracy, economy, journalism and freedom of expression. Designing and using automatic methods to detect fake news on social media has become a significant challenge. In this paper, we examine the publishers' role in detecting fake news on social media. We also suggest a high accurate multi-modal framework, namely FR-Detect, using user-related and content-related features with early detection capability. For this purpose, two new user-related features, namely Activity Credibility and Influence, have been introduced for publishers. Furthermore, a sentence-level convolutional neural network is provided to combine these features with latent textual content features properly. Experimental results have shown that the publishers' features can improve the performance of content-based models by up to 13% and 29% in accuracy and F1-score, respectively.