BaIT: Barometer for Information Trustworthiness

TL;DR

This paper introduces BaIT, a new approach for fake news classification that leverages pre-trained NLP models and data augmentation techniques, notably a novel sentence negation algorithm, to improve accuracy on under-represented classes.

Contribution

It proposes a novel sentence negation algorithm and demonstrates how pre-trained encoder models can be effectively used for fake news classification, especially addressing class imbalance.

Findings

Achieved comparable overall performance to existing baselines.

Significantly improved accuracy on under-represented classes.

Explored effective data augmentation methods for class imbalance.

Abstract

This paper presents a new approach to the FNC-1 fake news classification task which involves employing pre-trained encoder models from similar NLP tasks, namely sentence similarity and natural language inference, and two neural network architectures using this approach are proposed. Methods in data augmentation are explored as a means of tackling class imbalance in the dataset, employing common pre-existing methods and proposing a method for sample generation in the under-represented class using a novel sentence negation algorithm. Comparable overall performance with existing baselines is achieved, while significantly increasing accuracy on an under-represented but nonetheless important class for FNC-1.