Exploring the achievements and forecasting of SDG 3 using machine learning algorithms: Bangladesh perspective

Area of Science:

• Public Health
• Health Informatics
• Sustainable Development Goals

Background:

• Sustainable Development Goal 3 (SDG 3) is crucial for Bangladesh, focusing on health and well-being.
• Key indicators for SDG 3 progress in Bangladesh include Neonatal Mortality Rate (NMR), Under-5 Mortality Rate (U5MR), Maternal Mortality Ratio (MMR), and Death Rate Due to Road Traffic Injuries (RTI).

Purpose of the Study:

• To forecast key SDG 3 indicators in Bangladesh up to the year 2030.
• To compare these forecasts against the established 2030 SDG 3 targets.

Main Methods:

• Time series models, specifically Autoregressive Integrated Moving Average (ARIMA), were utilized for forecasting.
• Machine Learning (ML) models, including Bidirectional Recurrent Neural Networks (BRNN) and Elastic Neural Networks (ENET), were employed for enhanced predictive accuracy.
• Model selection was based on minimizing the Akaike Information Criterion (AIC) for ARIMA and evaluating error metrics like Root Mean Absolute Error (RMAE) and Root Mean Square Error (RMSE) for ML models.

Main Results:

• Elastic Neural Networks (ENET) demonstrated superior performance over BRNN and ARIMA models for NMR, U5MR, MMR, and Death Rate Due to RTI.
• Forecasted NMR and U5MR values are projected to fall below their respective 2030 SDG 3 targets, indicating positive progress.
• Forecasted MMR and Death Rate Due to RTI values are projected to exceed their 2030 SDG 3 targets, highlighting areas requiring focused intervention.

Conclusions:

• Bangladesh shows promising progress towards 2030 targets for NMR and U5MR.
• Addressing challenges in MMR and Death Rate Due to RTI is essential for achieving overall SDG 3 goals by 2030.
• The study provides critical insights for health policy and sustainable development in Bangladesh.