ExplainabilityTransparencyAccountabilitycs.CL
Explainable AI for Sentiment Analysis of Human Metapneumovirus (HMPV) Using XLNet
Md. Shahriar Hossain Apu, Md Saiful Islam, Tanjim Taharat Aurpa
February 1, 2025
arXiv.org
8 citations
Key Takeaways
-
AI can accurately gauge public health concerns: The study achieved 93.50% accuracy using XLNet to classify public sentiment about HMPV on social media, demonstrating that machine learning models can reliably capture how people feel about emerging health threats.
-
Transparency matters in health-related AI: The researchers used SHAP (SHapley Additive exPlanations) to make their sentiment analysis model explainable, showing which specific words and phrases drove the AI's classifications—an important step for building public trust in AI systems used for health monitoring.
-
Social media sentiment reflects real public health priorities: By analyzing sentiment around HMPV, the paper demonstrates that social media data can serve as an early indicator of public concern about disease outbreaks, potentially helping health authorities understand and address community anxieties more effectively.
-
Vulnerable populations drive public discourse: The research highlights that public concern about HMPV was particularly tied to its effects on vulnerable individuals, suggesting that AI sentiment analysis can help identify which populations are most on people's minds during health crises.
Abstract
In 2024, the outbreak of Human Metapneumovirus (HMPV) in China, which later spread to the UK and other countries, raised significant public concern. While HMPV typically causes mild symptoms, its effects on vulnerable individuals prompted health authorities to emphasize preventive measures. This paper explores how sentiment analysis can enhance our understanding of public reactions to HMPV by analyzing social media data. We apply transformer models, particularly XLNet, achieving 93.50% accuracy in sentiment classification. Additionally, we use explainable AI (XAI) through SHAP to improve model transparency.