Developing a Web Application to present future Dengue Risk maps based on machine learning models

Abstract

Effecient administration and control of public health, dengue epidemic prediction is essential. Conventional techniques don't produce timely or precise predictions; instead, they rely on environmental conditions and past data. This research suggests a dengue prediction system that forecasts epidemics by utilizing real-time data analysis and machine learning techniques. Through the integration of data from several sources, including demographics, weather patterns, and past dengue outbreaks, the system seeks to deliver timely insights for resource allocation and preventive measures. The technical part entails creating prediction models with random forest and ARIMA algorithms, trained on historical dengue data from Kaggle between 2013 and 2021. Metrics like accuracy, precision, and recall will be used to assess the system's performance. The suggested method for predicting dengue fever has the potential to revolutionize public health strategies by providing accurate forecasts and actionable insights for dengue prevention and control.