Label Noise-Resistant Mean Teaching for Weakly Supervised Fake News Detection

TL;DR

This paper introduces LNMT, a novel weakly supervised fake news detection method that leverages unlabeled data and user comments, employing a mean teacher framework to refine labels and mitigate noise, resulting in improved detection accuracy.

Contribution

The paper proposes a label noise-resistant mean teaching approach (LNMT) that effectively utilizes unlabeled data and feedback comments for weakly supervised fake news detection, with mechanisms to suppress label noise.

Findings

LNMT outperforms existing methods in fake news detection accuracy.

The mean teacher framework effectively refines weak labels and reduces noise impact.

Extensive experiments validate the superior performance of LNMT.

Abstract

Fake news spreads at an unprecedented speed, reaches global audiences and poses huge risks to users and communities. Most existing fake news detection algorithms focus on building supervised training models on a large amount of manually labeled data, which is expensive to acquire or often unavailable. In this work, we propose a novel label noise-resistant mean teaching approach (LNMT) for weakly supervised fake news detection. LNMT leverages unlabeled news and feedback comments of users to enlarge the amount of training data and facilitates model training by generating refined labels as weak supervision. Specifically, LNMT automatically assigns initial weak labels to unlabeled samples based on semantic correlation and emotional association between news content and the comments. Moreover, in order to suppress the noises in weak labels, LNMT establishes a mean teacher framework equipped with label propagation and label reliability estimation. The framework measures a weak label similarity matrix between the teacher and student networks, and propagates different valuable weak label information to refine the weak labels. Meanwhile, it exploits the consistency between the output class likelihood vectors of the two networks to evaluate the reliability of the weak labels and incorporates the reliability into model optimization to alleviate the negative effect of noisy weak labels. Extensive experiments show the superior performance of LNMT.