Pandemic coronavirus disease (Covid-19): World effects analysis and prediction using machine-learning techniques

Area of Science:

• Epidemiology
• Infectious Disease Modeling
• Artificial Intelligence in Public Health

Background:

• The novel Coronavirus (COVID-19) pandemic, originating in Wuhan, China, has had a significant global impact.
• Existing predictive models struggle with data uncertainty and inaccuracies, hindering effective epidemic control.
• Accurate forecasting is crucial for informed decision-making and implementing appropriate public health measures.

Purpose of the Study:

• To develop and evaluate an Artificial Intelligence (AI)-based meta-analysis for predicting global COVID-19 epidemic trends.
• To address the limitations of traditional methods in forecasting pandemic trajectories.
• To provide insights for proactive governmental and citizen responses.

Main Methods:

• Application of machine learning algorithms: Naïve Bayes, Support Vector Machine (SVM), and Linear Regression.
• Utilized a real-time series dataset encompassing global confirmed, recovered, deaths, and active COVID-19 cases.
• Conducted statistical analysis on symptoms, top affected countries, and co-active cases.

Main Results:

• Naïve Bayes demonstrated superior performance in predicting COVID-19 future trends compared to SVM and Linear Regression.
• Lower Mean Absolute Error (MAE) and Mean Squared Error (MSE) values for Naïve Bayes indicate its effectiveness.
• Statistical analysis provided insights into disease patterns and country-specific impacts.

Conclusions:

• AI-based meta-analysis, particularly using Naïve Bayes, offers a promising approach for accurate COVID-19 trend prediction.
• The study establishes a benchmark for machine learning applications in outbreak forecasting.
• Findings support proactive global responses to the ongoing pandemic.