A Weighted Ensemble Model for Prediction of Dengue Occurrence in North India (Chandigarh)

Abstract

In tropical nations, dengue fever is one of the most widespread vector-borne infections, particularly in developing countries such as India, Bangladesh, and Pakistan. Dengue fever can range from mild to severe fever cases. Dengue fever is an epidemic spread by mosquitos that affects people of all ages in over a hundred countries throughout the world. The research examines real-time series prediction and analysis using three regression models, as well as the development of a weighted average prediction model for infectious illness prediction. The integrated diseases monitoring programme of the Government of India provided monthly statistics on dengue cases from 2014 to 2017. Three regression models were used to analyse data: support vector regression, neural network, and linear regression. Mean Absolute Error, Root Mean Square Error, and Mean Square Error are some of the performance criteria that have been employed. In terms of its effectiveness, it was discovered that the postulated weighted ensemble model performed better. The primary purpose of this project is to reduce prediction errors, and we discovered that our planned weighted ensemble model is more effective in this regard.