Support Vector Machine Performance In Diabetes Data Classification With GridsearchCV Optimization

Sparta, I Wayan Adi
Parianthana, Putu Eka

Jurnal Multidisiplin Sahombu
Vol. 5 No. 01 (2025): Jurnal Multidisiplin Sahombu (2025)

Diabetes is one of the most common chronic diseases and has the potential to cause serious complications if not detected early. This study evaluates the performance of the Support Vector Machine (SVM) algorithm in diabetes data classification using parameter optimization with GridSearchCV. The dataset used includes eight health features from Kaggle, which include factors such as glucose levels, blood pressure, and body mass index (BMI). After parameter optimization, the SVM model achieved 74% accuracy on the test data, with precision and recall performance varying between classes. This research highlights the importance of hyperparameter optimization in improving the accuracy and balance of health data classification. These results provide insight into the application of machine learning technologies to support early detection and medical decision-making.