Oleh
Universitas Raharja

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Detecting Post-Traumatic Stress Disorder Trauma with Hybrid Expert System and Machine Learning Sekar; Oleh; Sutrisno
CCIT Journal Vol 19 No 2 (2026): CCIT JOURNAL
Publisher : UNIVERSITAS RAHARJA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33050/ccit.v19i2.4379

Abstract

Post-Traumatic Stress Disorder (PTSD) remains a critical mental health challenge, yet conventional diagnostics often struggle with subjective bias and linguistic ambiguity. This research addresses these limitations by developing a Hybrid Expert System that integrates Forward Chaining with a Random Forest machine learning algorithm to classify PTSD trauma severity into three levels: Low, Medium, and High. The hybrid architecture combines rigid clinical logic with adaptive pattern recognition to better interpret complex patient narratives. To overcome the prevalent issue of class imbalance in medical data, the Synthetic Minority Over-sampling Technique (SMOTE) was employed. Experimental results demonstrate that the proposed system achieves a high accuracy of 99.02%, significantly outperforming traditional keyword-matching methods and standalone rule-based systems. While maintaining high precision, the integration of SMOTE successfully enhanced the model’s sensitivity in detecting severe trauma cases. This study underscores the importance of hybrid intelligence in mental health informatics, providing a more objective, transparent, and reliable diagnostic tool. Such technology is vital for early clinical intervention, potentially reducing the long-term socio-economic impact of undiagnosed psychological trauma and supporting medical professionals in delivering more accurate mental health services.