Development of machine learning modelling and dengue risk mapping: a concept framework

Abstract

Dengue fever (DF) has emerged as the world most common mosquito-borne diseases. DF has more than tripled in the last five decades. The disease is primarily present in tropical and subtropical areas, putting around one-third of the world’s population at risk of infection. As a result of rising urbanization, broad global travel, a lack of sufficient mosquito control measures, and globalization, dengue viruses have spread rapidly over the world. Climate variables influence DF incidence and fatality rates, but so do sociodemographic factors, rendering certain demographic subpopulations more vulnerable to infection. However, due to the methodological difficulties associated with integrating different data sources, only a few studies addressing the causes of dengue incidence incorporate both meteorological and sociodemographic components. The application of the Geographical Information System (GIS) and Machine Learning (ML) Algorithm will be introduced in this study to act as a crucial tool in the epidemiological and spatial investigation, and the machine learning algorithm will act as a tool to analyze and make predictions based on data, without being explicitly programmed to do so. The study’s findings are supposed to aid Malaysian dengue management efforts. Although the study framework was established for DF, it can be adapted to incorporate other mosquito-borne diseases such as Malaria and Chikungunya, as well as non-mosquito-borne problems.