Stunting is a chronic nutritional problem affecting the physical growth and cognitive development of toddlers, especially in early childhood. This study employs the K-Nearest Neighbors (K-NN) algorithm to determine stunting status based on anthropometric variables such as age, weight, and height. The algorithm categorizes data using proximity between samples. Data from the Salatiga City Health Department in 2024 were normalized and encoded for analysis. K-NN was chosen for its ability to provide high-accuracy nutritional classification. Results show that the algorithm achieved 100% accuracy, precision, and recall at certain K values, particularly in the small K range (2-8), demonstrating its effectiveness in identifying nutritional status in toddlers. This study is expected to serve as a reference for utilizing technology and data analysis in early stunting detection, aiding healthcare professionals in designing more targeted and effective interventions. Additionally, it opens opportunities for further development to enhance diagnostic accuracy using machine learning technology.
Read full abstract