A Review of Epidemic Forecasting Using Artificial Neural Networks

Abstract

Background and aims: Since accurate forecasts help inform decisions for preventive health-care<br />intervention and epidemic control, this goal can only be achieved by making use of appropriate<br />techniques and methodologies. As much as forecast precision is important, methods and model<br />selection procedures are critical to forecast precision. This study aimed at providing an overview of<br />the selection of the right artificial neural network (ANN) methodology for the epidemic forecasts. It is<br />necessary for forecasters to apply the right tools for the epidemic forecasts with high precision.<br />Methods: It involved sampling and survey of epidemic forecasts based on ANN. A comparison of<br />performance using ANN forecast and other methods was reviewed. Hybrids of a neural network with<br />other classical methods or meta-heuristics that improved performance of epidemic forecasts were<br />analysed.<br />Results: Implementing hybrid ANN using data transformation techniques based on improved<br />algorithms, combining forecast models, and using technological platforms enhance the learning and<br />generalization of ANN in forecasting epidemics.<br />Conclusion: The selection of forecasting tool is critical to the precision of epidemic forecast; hence, a<br />working guide for the choice of appropriate tools will help reduce inconsistency and imprecision in<br />forecasting epidemic size in populations. ANN hybrids that combined other algorithms and models,<br />data transformation and technology should be used for an epidemic forecast.