A probe into performance analysis of real-time forecasting of endemic infectious diseases using machine learning and deep learning algorithms

Abstract

The current work aims at probing the performance of real-time forecasting of endemic infectious diseases by means of machine learning and deep learning techniques. An LSTM-based time series forecasting framework and machine learning-based framework are proposed for forecasting the endemic infectious diseases in real time. With recent outbreaks of Ebola, Zika, cholera, and COVID 2019, a question is being raised on our alertness as well as preparedness toward controlling the spread of these pandemics. Accurate and reliable prediction occurrences of these diseases are compulsory for the health personals to enable timely response in handling these outbreaks. The diversities of the communities make it more complex along with the humongous data generated due to the convergence of SMAC technologies. The data generated due to this complex network is nonlinear and non-stationary. Processing of this data requires an effort from a multidimensional perspective. The current work proposed the utilization of machine learning and deep learning-based long short-term memory (LSTM) techniques for the assessment of time series forecasting of casualties in case of cholera outbreak that happened recently in Yemen. The feasibility of these two techniques is probed using performance evaluation metrics. The core objective of using these two techniques is in considering nonlinear and non-stationary behavior. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.