Leveraging the power of transformers for guilt detection in text

TL;DR

This paper tackles guilt detection in text, an underexplored emotion in NLP, using transformer-based language models. It introduces GuiltBERT, a BERT-based model finetuned on guilt data, and systematically compares it to BERT and RoBERTa on ISEAR and VIC datasets for both multiclass emotion detection and binary guilt classification. The results show that GuiltBERT achieves competitive general-emotion performance and state-of-the-art guilt detection on VIC, with some confusion among guilt, disgust, and shame. The work provides a pathway toward more nuanced emotion understanding in NLP and suggests directions for architecture scaling and explainable guilt analysis.

Abstract

In recent years, language models and deep learning techniques have revolutionized natural language processing tasks, including emotion detection. However, the specific emotion of guilt has received limited attention in this field. In this research, we explore the applicability of three transformer-based language models for detecting guilt in text and compare their performance for general emotion detection and guilt detection. Our proposed model outformed BERT and RoBERTa models by two and one points respectively. Additionally, we analyze the challenges in developing accurate guilt-detection models and evaluate our model's effectiveness in detecting related emotions like "shame" through qualitative analysis of results.

Figures (3)

• Figure 1: Confusion Matrix of GuiltBERT Model in the ISEAR Dataset
• Figure 2: Confusion Matrix of GuiltBERT Model in the VIC balanced Dataset
• Figure 3: Confusion Matrix of GuiltBERT Model in the VIC binary Dataset