GWO-SVM: AN APPROACH TO IMPROVING SVM PERFORMANCE USING GREY WOLF OPTIMIZER IN INTELLECTUAL DISABILITY CLASSIFICATION

Intellectual disability (ID) is a neurodevelopmental disorder that requires early and accurate diagnosis. This study aims to improve the efficiency of ID diagnosis using a machine learning approach. A Support Vector Machine (SVM) model optimized with Grey Wolf Optimizer (GWO) was developed and trained using data from questionnaires completed by 101 families/guardians of ID patients at RSUD Dr. Soetomo Surabaya. The features used include family history, cognitive abilities, and adaptive behaviors. The results showed that the GWO-SVM model achieved an accuracy of 95% in classifying ID patients, an improvement of 5% compared to the conventional SVM. The GWO algorithm successfully optimized the parameters in SVM, resulting in a model with the best performance. These findings indicate the potential of GWO-SVM as an effective and efficient tool for assisting in the diagnosis of ID.