Deep Learning for Virus-Spreading Forecasting: a Brief Survey

TL;DR

This survey reviews deep learning methods for predicting virus spread, highlighting emerging trends, comparing classical and hybrid models, and discussing their advantages, disadvantages, and future research directions.

Contribution

It provides a comprehensive overview of deep learning approaches for virus-spreading forecasting, emphasizing recent developments and potential strategies.

Findings

Classical deep learning models are effective for spatial-temporal prediction.

Hybrid models combine multiple techniques for improved accuracy.

Identifies key challenges and future directions in virus-spreading forecasting.

Abstract

The advent of the coronavirus pandemic has sparked the interest in predictive models capable of forecasting virus-spreading, especially for boosting and supporting decision-making processes. In this paper, we will outline the main Deep Learning approaches aimed at predicting the spreading of a disease in space and time. The aim is to show the emerging trends in this area of research and provide a general perspective on the possible strategies to approach this problem. In doing so, we will mainly focus on two macro-categories: classical Deep Learning approaches and Hybrid models. Finally, we will discuss the main advantages and disadvantages of different models, and underline the most promising development directions to improve these approaches.