An approach to forecast impact of Covid-19 using supervised machine learning model

Area of Science:

• Epidemiology
• Data Science
• Public Health

Background:

• The Covid-19 pandemic highlighted the critical need for effective disease forecasting.
• Traditional forecasting methods are often short-term and country-specific.
• Robust techniques are essential for disease detection, alleviation, and prevention.

Purpose of the Study:

• To propose a novel multimodel machine learning technique for long-term Covid-19 parameter forecasting.
• To provide accurate predictions for India and on a global scale.
• To address limitations of existing short-term and country-specific forecasting models.

Main Methods:

• Development of the EAMA (Epidemiological Analysis and Modeling Approach) hybrid model.
• Utilizing two distinct datasets: Ministry of Health & Family Welfare (India) and Worldometers.
• Application of machine learning for analyzing past and present data to predict future trends.

Main Results:

• The EAMA model demonstrated high accuracy in long-term forecasting of Covid-19 parameters.
• Predicted data closely aligned with real-time values for both India and global trends.
• Successful state-wise and country-wise prediction experiments were conducted.

Conclusions:

• The EAMA hybrid model offers a robust solution for long-term epidemiological forecasting.
• This technique can significantly support public health initiatives in managing pandemics.
• Accurate, long-term forecasting is crucial for timely intervention and mitigation efforts.