<![CDATA[Stunting is a significant public health problem, impacting the physical and cognitive growth and development of children under five. In Indonesia, stunting is a major issue caused by a lack of nutritional intake since birth, including in the city of Semarang. This study aims to compare the performance of K-Nearest Neighbor (KNN), Naïve Bayes, and Gradient Boosting Classifier algorithms in classifying stunting in toddlers, to find the best model. The data used is quantitative data from posyandu, consisting of 1288 samples with variables including Name, Gender, Age, Date of Birth, Parent's Name, Village, Rt, RW, weight, height, arm circumference, and Z-score. After data collection, a data preprocessing process is carried out to clean and prepare the data. The data was divided into training and test data with a ratio of 80:20, 70:30, and 60:40, which were then trained and tested using the three algorithms. The best model was further evaluated with K-Fold Cross Validation to assess the stability and generalizability of the predictions. Model evaluation uses accuracy, precision, recall and F1-Score metrics. The results showed that Gradient Boosting Classifier gave the best performance with 99.92% accuracy, 99.92% precision, 99.92% recall, and 99.92% F1-score. This study concludes that the Gradient Boosting Classifier is the most optimal model in the classification of stunting in toddlers, giving the best precision results.]]>
Copyrights © 2024
