Multimodal Fake News Detection

TL;DR

This paper presents a multimodal fake news detection method combining text and images using CNNs, achieving higher accuracy than text-only models, and performs fine-grained classification on the Fakeddit dataset.

Contribution

It introduces a multimodal approach for fine-grained fake news classification that outperforms unimodal text-only methods, demonstrating the value of combining text and images.

Findings

Multimodal CNN approach achieves 87% accuracy.

Images significantly improve detection of manipulated and satirical news.

Text-only BERT model achieves 78% accuracy.

Abstract

Over the last years, there has been an unprecedented proliferation of fake news. As a consequence, we are more susceptible to the pernicious impact that misinformation and disinformation spreading can have in different segments of our society. Thus, the development of tools for automatic detection of fake news plays and important role in the prevention of its negative effects. Most attempts to detect and classify false content focus only on using textual information. Multimodal approaches are less frequent and they typically classify news either as true or fake. In this work, we perform a fine-grained classification of fake news on the Fakeddit dataset, using both unimodal and multimodal approaches. Our experiments show that the multimodal approach based on a Convolutional Neural Network (CNN) architecture combining text and image data achieves the best results, with an accuracy of 87%. Some fake news categories such as Manipulated content, Satire or False connection strongly benefit from the use of images. Using images also improves the results of the other categories, but with less impact. Regarding the unimodal approaches using only text, Bidirectional Encoder Representations from Transformers (BERT) is the best model with an accuracy of 78%. Therefore, exploiting both text and image data significantly improves the performance of fake news detection.