Statistical and network-based analysis of Italian COVID-19 data: communities detection and temporal evolution

Marianna Milano; Mario Cannataro

In Italy, COVID-19 has been identified on January 2020 [5] and the outbreak started in Lombardia and Veneto at the end of February 2020. From the northern regions of Italy, the disease spread very quickly to the nearest regions and then to the rest ones. Italy has considered one of the main epicentres of the pandemic recording a number of infected equal to 97689 and a number of deaths equal to 10799 up to 29rd of March. In this paper we want to underline difference at regional level and to evidence how the disease has spread along time. To do this, both statistical and network-based analysis on Italian COVID-19 data has been performed. The data are daily provided by the Italian Civil Protection, and they refer to the period February 24 to March 29, 2020. In particular, we analyzed the Italian COVID-19 data at the regional level in order to evaluate which regions show a similar trend. Then, for each type of data, we built the related networks where the nodes represent the Italian region, and the edges represent a level of similarity among them. Finally, we evaluate how the networks evolve over the weeks up to the end of March 2020. The main contribution of the the paper is a network-based representation of COVID-19 diffusion similarity among regions and graphbased visualization to underline similar diffusion regions. The rest of the paper is organized as follows: Section 2 discusses the Italian COVID-19 data, the statistical and network-based analysis, Section 3 discusses the results, Finally, Section 4 concludes the paper.