Understanding Public Perception of Crime in Bangladesh: A Transformer-Based Approach with Explainability
This work addresses the need for analyzing public perception of crime in a low-resource language context, with potential applications in policy-making, though it is incremental in applying known methods to a new domain.
The study tackled the problem of classifying public sentiment on crime-related news in Bangladesh by developing a transformer-based model that achieved 97% accuracy on a new Bangla-language dataset of 28,528 social media comments, outperforming existing methods.
In recent years, social media platforms have become prominent spaces for individuals to express their opinions on ongoing events, including criminal incidents. As a result, public sentiment can shift dynamically over time. This study investigates the evolving public perception of crime-related news by classifying user-generated comments into three categories: positive, negative, and neutral. A newly curated dataset comprising 28,528 Bangla-language social media comments was developed for this purpose. We propose a transformer-based model utilizing the XLM-RoBERTa Base architecture, which achieves a classification accuracy of 97%, outperforming existing state-of-the-art methods in Bangla sentiment analysis. To enhance model interpretability, explainable AI technique is employed to identify the most influential features driving sentiment classification. The results underscore the effectiveness of transformer-based models in processing low-resource languages such as Bengali and demonstrate their potential to extract actionable insights that can support public policy formulation and crime prevention strategies.