A Weighted Ensemble Model for Prediction of Dengue Incidence in North India (Chandigarh)
Dengue is classified as an extremely epidemic among all of the vector-borne diseases (VBDs) of tropical states especially the developing countries like India, Bangladesh Pakistan, etc. The dengue diseases are varying from mild fever to dengue homographic fever. Dengue infections effects are more than a hundred countries worldwide including individuals of all age groups (infants, children adolescents, and adults). The paper deals with the real-time series prediction and analysis by three regression models as well the development of the weighted average ensemble model meant for the forecast of the deadly catching diseases. The Regular data on dengue virus prevalence for several years (2014-2017) were engaged from various integrated programs of diseases surveillance by (nvbdcp) the Regime of India. The statistics were analyzed by tierce regression models which are the neural network, support vector regression, and linear regressions. Basically, we used here different performance parameters such as Mean Absolute Error, Root Mean Square Error, and the Mean Square Error, and it was initiated that our proposed model is weighted ensemble model will work better in standings of performance measures. Basically, the foremost purpose of this study is to reduce the error caused during the forecast; our anticipated model performed better in terms of forecast errors.
Keywords – Weighted Ensemble, Prediction, Dengue, Regression