International Conference on Computer and Knowledge Engineering

Home / 11th International Conference on Computer and Knowledge Engineering

A novel hybrid DMHS-GMDH algorithm to predict COVID-19 pandemic time series

Authors :

Ahmad Taheri¹ Shahriar Ghashghaei² Amin Beheshti³ Keyvan RahimiZadeh⁴

1- Yasouj University 2- Tehran University of Medical Sciences 3- Macquarie University 4- Yasouj University, Macquarie University

Keywords :

COVID-19, pandemic, machine learning, harmony search, metaheuristic algorithm

Abstract :

In this paper, a novel hybrid method called DMHS-GMDH is presented to predict the time series of COVID-19 outbreaks. In this way, a new version of Harmony Search (HS) algorithm, named Double Memory HS (DMHS), is designed to optimize the structure of a Group Method of Data Handling (GMDH) type neural network. We conduct a series of experiments by applying proposed method on real COVID-19 dataset to forecast new cases and deaths of COVID-19. The statistical analysis indicates that the DMHS-GMDH algorithm on average provides better results than other competitors and the results demonstrate how our approach at least improves coefficient of determination and RMSE by 21% and 45%, respectively.