Fake News Detection System using XLNet model with Topic Distributions: CONSTRAINT@AAAI2021 Shared Task

TL;DR

This paper presents a fake news detection system that combines XLNet language model representations with topic distributions from LDA, achieving high accuracy in COVID-19 fake news detection.

Contribution

The novel approach integrates XLNet with LDA topic distributions for improved fake news detection performance.

Findings

XLNet + Topic Distributions outperforms baselines with an F1-score of 0.967.

The method effectively captures contextual and topical information for fake news detection.

Comparison shows significant improvement over existing methods.

Abstract

With the ease of access to information, and its rapid dissemination over the internet (both velocity and volume), it has become challenging to filter out truthful information from fake ones. The research community is now faced with the task of automatic detection of fake news, which carries real-world socio-political impact. One such research contribution came in the form of the Constraint@AAA12021 Shared Task on COVID19 Fake News Detection in English. In this paper, we shed light on a novel method we proposed as a part of this shared task. Our team introduced an approach to combine topical distributions from Latent Dirichlet Allocation (LDA) with contextualized representations from XLNet. We also compared our method with existing baselines to show that XLNet + Topic Distributions outperforms other approaches by attaining an F1-score of 0.967.