Understanding Aviation Mental Health with Explainable Artificial Intelligence in Incident Reports

Abstract

Explainable Artificial Intelligence (XAI) is a flourishing field that extends beyond mere predictions to clarify complex paradigms in aviation mental health, offering insights into the traces of mental health-related cases in aviation incidents. This study enriches the discourse by examining aviation professionals' psychological well-being and stress factors, deploying Text Mining, Information Fusion, SHAP, and LIME XAI methods. These innovative approaches provide transparent, interpretable models that enhance understanding of the traces of the mental health challenges in the aviation industry, potentially guiding interventions, and support systems for improved mental health outcomes. The mental health-related issues are Tabu in the aviation community that are also hidden in the incident reports. This study will investigate the patterns that create mental health-related incidents in FAA reports using XAI methods.

Author

• Cankaya, Mehmet Burak Embry-Riddle Aeronautical University

Keywords

• XAI
• Explainable Artificial Intelligence
• Text Ming
• Aviation Safety
• Aviation Mental Health
• Advanced Air Mobility
• Business Analytics
• Business Intelligence
• Data Science
• Psychology