Hidayati, Rahmatina and Anita, Anita and Lestanti, Sri (2025) Evaluation of Distance Measurement Techniques in the k-NN Method for Toddler Nutritional Status Classification. Journal of Information System and Application Development, 3 (1). pp. 31-37. ISSN P-ISSN: 2988-5698; E-ISSN: 2988-4721

Preview	Text Evaluation of Distance Measurement Techniques in the k-NN Method for Toddler Nutritional Status Classification.pdf Download (410kB) | Preview
Preview	Text Hasil Cek Plagiasi_Evaluation of Distance Measurement Techniques in the k-NN Method.pdf Download (1MB) | Preview

Abstract

Toddler nutritional status is an essential indicator in assessing public welfare and health. At Rongga Koe Village Health Post, determining nutritional status is still done manually, so it take sa long time and is prone to errors. This study aims to develop a classification system for toddler nutritional status using the K-Nearest Neighbors (k-NN). The data used was 100 samples with five parameters: gender, age, weight, height, and upper arm circumference. The classification process was carried out with variations in the ratio of training and testing data (90:10, 80:20, 70:30, 60:40), as well as the k value and distance calculation method (Euclidean, Manhattan, Chebyshev, Mahalanobis). The results showed that the best combination was obtained at a ratio of 90:10 and a k value = 9 with the Mahalanobis Distance method, which achieved the highest accuracy of 85.7%. This study proves that the K-NN method is effective in helping to classify nutritional status digitally and more efficiently.

Actions (login required)

View Item