<![CDATA[Gallstone disease is a common medical condition that often presents without symptoms until complications occur. Early prediction of this disease can improve patient outcomes through timely intervention. This study aims to compare the performance of five classification algorithms in predicting gallstone disease using clinical data: Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbors (K-NN). The dataset used was obtained from the UCI Machine Learning Repository and contains clinical data from 319 patients, comprising 38 numerical features. Outlier handling was conducted using Winsorized Transformation and RobustScaler. Each model was optimized using GridSearchCV and evaluated using accuracy, precision, recall, F1-score, and ROC AUC metrics. The results show that SVM and Logistic Regression achieved the best performance, each with 85.9% accuracy and an AUC above 0.90. Based on these findings, Logistic Regression and SVM are recommended as the most effective classification models for gallstone disease prediction using clinical data]]>
Copyrights © 2026
