Penerapan Algoritma Decision Tree, Svm, Naive Bayes Dalam Deteksi Stunting Pada Balita

Stunting is a toddler's body condition that is short according to body length according to age (PB/U), ≤ 2 Standard Deviations (SD), with a z-score between -3 standard deviations (SD). Where checking the stunting status of toddlers takes quite a long time because it is done manually and is also prone to errors. Therefore, it is hoped that a system can classify toddler examination data quickly and accurately to predict children's stunting status. Building a system that uses an algorithm to classify the stunting status of toddlers using a decision tree, naïve Bayes, and SVM. With what level of accuracy is the best of the 3 algorithms? Results from testing with 30% testing data and 70% training data using an algorithm decision tree, naïve Bayes, and SVM. Accuracy level test results decision tree by 99%, naïve Bayes of 48%, and SVM of 95%. So, the algorithm with the highest level of accuracy is the decision tree amounts to 99%. Wallet hire decision tree is better for detecting stunting in toddlers.