Sistem Prediksi Pertumbuhan Balita Menggunakan Metode Prophet Berdasarkan Data Historis Pertumbuhan

Sudarmanto, Audrey Pramudita (2026) Sistem Prediksi Pertumbuhan Balita Menggunakan Metode Prophet Berdasarkan Data Historis Pertumbuhan. Undergraduate thesis, Politeknik Negeri Jember.

	Text (Abstract) Abstract.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (310kB)
	Text (Bab 1 Pendahuluan) Bab 1 Pendahuluan.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (298kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (278kB)
	Text (Laporan Lengkap) Laporan Lengkap.pdf - Submitted Version Restricted to Registered users only Download (8MB) | Request a copy

Abstract

Monitoring toddler growth is a critical effort in preventing chronic nutritional problems such as stunting, which remains a significant health challenge in Indonesia, yet current monitoring at community health posts (Posyandu) is conducted manually and reactively, resulting in delayed nutritional interventions. This study designed and implemented a toddler growth prediction system based on the Prophet method, integrated into an Android application named Kawan Tumbuh, developed using the Flutter framework with a Python (Flask) backend, utilizing 259 historical records of Posyandu toddlers aged 24–60 months, configured with a growth=logistic parameter consistent with the biological growth pattern of toddlers, capable of predicting three key anthropometric indicators — weight, height, and head circumference — for 1–3 months ahead along with nutritional status classification based on Z-Score according to WHO and Indonesian Ministry of Health standards. Model evaluation results demonstrated excellent performance: weight achieved MAE of 0.309 kg, RMSE of 0.339 kg, and MAPE of 1.88% (accuracy 98.12%); height achieved MAE of 0.285 cm, RMSE of 0.320 cm, and MAPE of 0.27% (accuracy 99.73%); and head circumference achieved MAE of 0.131 cm, RMSE of 0.140 cm, and MAPE of 0.27% (accuracy 99.73%). Expert validation using an ISO/IEC 25010-based instrument yielded a feasibility score of 89.29% classified as Highly Feasible, while Black-Box Testing on 20 toddler samples produced a nutritional status classification accuracy of 90.00% against expert diagnosis as the gold standard, demonstrating that the system is effective and feasible for implementation as an Android-based nutritional monitoring tool in Posyandu settings

Actions (login required)

View Item