<![CDATA[This study aims to compare the performance of K-Nearest Neighbors (KNN), Artificial Neural Network (ANN), and Support Vector Machines (SVM), classification methods to find the best suited method to train a machine to classify someone to their group of obesity levels according to their eating habits and physical condition. This experiment uses the “Estimation of Obesity Levels based on Eating Habits and Physical Condition” dataset. The primary focus is on achieving high accuracy score and complex decision boundaries handling without minding the long training times, considering misclassification in the medical field might cause fatal consequences. This experiment’s result shows that the SVM classification method with linear kernel provides the best overall performance for classifying obesity level, with the average accuracy of 0.944, precision of 0.944, recall of 0.942, and f1-score of 0.942. Notably, with the help of C kernel parameter of 200, the model teaches near-perfect performance evaluation scores that has the result of 0.99 score in accuracy, precision, recall, and f1-score.]]>
Copyrights © 2025
