DESIGN OF A WEB APPLICATION FOR RISK OF MALARIA PREDICTION USING ENVIRONMENTAL DATA

Malaria remains one of the most significant public health challenges in sub-Saharan Africa, causing millions of infections and deaths annually, particularly in vulnerable communities. The transmission of malaria is closely linked to environmental factors such as temperature, humidity, and rainfall, which directly influence mosquito breeding and mortality. Traditional methods of Page | 5 malaria prevention remain largely reactive, with limited emphasis on proactive prediction. This research demonstrated the conceptualization and realization of a mobile application designed to predict malaria risk levels using environmental data and machine learning techniques. Historical weather data and malaria case records from sampled Nigerian cities were collected, preprocessed, and applied to train a Random Forest algorithm. The trained algorithm was deployed in a web application featuring a user-friendly interface. Users can either manually input environmental variables or allow the app to automatically retrieve weather data based on their location. The interactive interface provides malaria risk predictions categorized as High, Moderate, or Low, alongside tailored recommendations for prevention and educational information about malaria. The tool exhibited high precision in validation and provided an affordable, user-friendly tool for individuals, clinicians, and decision-makers to adopt prompt preventive measures. The tool can be transformed in the future through the incorporation of other predictive variables, become accessible in offline environments, and be incorporated into the healthcare system to greatly benefit malaria preventive measures, especially in rural, resource-scarce environments. Keywords: Health Informatics, Malaria Prediction, Machine Learning, Risk Assessment, Environmental Data