Explainable Verbal Deception Detection using Transformers

TL;DR

This paper explores transformer-based deep learning models for automated deception detection in text, emphasizing interpretability through explanations and linguistic feature analysis, achieving improved accuracy and insights into deception cues.

Contribution

It introduces and evaluates six transformer-based models for deception detection, incorporating interpretability methods like LIME and linguistic analysis, which is a novel approach in this domain.

Findings

Transformer models improve detection accuracy by 2.11%.

LIWC features differ significantly between truthful and deceptive texts.

Model explanations reveal linguistic cues associated with deception.

Abstract

People are regularly confronted with potentially deceptive statements (e.g., fake news, misleading product reviews, or lies about activities). Only few works on automated text-based deception detection have exploited the potential of deep learning approaches. A critique of deep-learning methods is their lack of interpretability, preventing us from understanding the underlying (linguistic) mechanisms involved in deception. However, recent advancements have made it possible to explain some aspects of such models. This paper proposes and evaluates six deep-learning models, including combinations of BERT (and RoBERTa), MultiHead Attention, co-attentions, and transformers. To understand how the models reach their decisions, we then examine the model's predictions with LIME. We then zoom in on vocabulary uniqueness and the correlation of LIWC categories with the outcome class (truthful vs deceptive). The findings suggest that our transformer-based models can enhance automated deception detection performances (+2.11% in accuracy) and show significant differences pertinent to the usage of LIWC features in truthful and deceptive statements.