ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM UNTUK PERAMALAN STUNTING DI ASIA TENGGARA TAHUN 1997 - 2022

Stunting is a condition of chronic malnutrition in children under five, representing a significant global health issue, particularly in Southeast Asia. This study aims to develop a forecasting model for stunting prevalence in Southeast Asia using the Adaptive Neuro-Fuzzy Inference System (ANFIS) method for the 1997-2022 period. ANFIS was chosen for its ability to combine fuzzy logic and neural networks to capture complex patterns in historical data. Secondary data on stunting prevalence were obtained from the World Health Organization (WHO), with the year 2022 as the cut-off point, as the latest data officially available on the WHO website only extend to that year. The research process involved dataset construction, the development of a fuzzy inference system, model training, and forecasting evaluation. The model evaluation was conducted using the metrics Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE) to measure forecasting accuracy. The results showed that the ANFIS model achieved a MAPE of 6.85%, an MSE of 0.1933 and an RMSE of 0.4396, indicating good forecasting accuracy. The comparison graph between actual data and forecasting results demonstrated a high level of agreement. These findings highlight the effectiveness of ANFIS in forecasting stunting prevalence, suggesting its potential as a valuable tool in health policy planning to address stunting in Southeast Asia. This study contributes to the development of more accurate predictive models, supporting more efficient health intervention planning.
Keywords: Adaptive Neuro-Fuzzy Inference System, Southeast Asia, Child Health, Forecasting, Stunting.