O Impacto da Pandemia do COVID-19 no Comportamento do Tráfego de Rede e no Processo de Predição

• Rafael A. Menezes UECE
• Dyego H. L. Oliveira UECE
• Rafael L. Gomes UECE

DOI: https://doi.org/10.5753/wpeif.2021.17196

Abstract

The COVID-19 pandemic brought with it numerous changes, from the economy to social behavior. This new reality caused a profound impact on the network traffic of universities, companies, institutions, etc. Similarly, the traffic prediction process was also impacted, since the behavior of elastic demand for resources was indirectly affected. Within this context, this work presents an experimental analysis of the network traffic behavior, as well as the resource demand prediction of real datasets in the pre-pandemic period and during the pandemic. The results of the experiments carried out show the importance of stationary data identification in order to adjust the time series, minimizing prediction errors.

References

Dokumentov, A. and Hyndman, R. J. (2015). STR: A Seasonal-Trend Decomposition Procedure Based on Regression. Monash Econometrics and Business Statistics Working Papers 13/15, Monash University, Department of Econometrics and Business Statistics.

Enders, W. (2015). Applied econometric time series. Wiley, 4 edition.

Grossi, M. G. R., Minoda, D. d. S. M., and FONSECA, R. G. P. (2020). Impacto da pandemia do covid-19 na educação: Reflexos na vida das famílias. Teoria e Prática da Educação, 23(3):150–170.

i Silvestre, J. L. C., i Rosselló, A. S., and Ortuño, M. A. (2001). Unit root and stationarity tests’ wedding. Economics Letters, 70(1):1–8.

Oliveira, D. H. L. (2020). Modelo adaptativo para previsão de demanda por recursos de rede em provedores de internet modernos.

Pedro Guilherme Costa Ferreira, Anna Carolina Barros, D. M. d. M. I. C. L. d. O. e. V. E. L. d. A. D. (2017). Análise de séries temporais em R: curso introdutório. Addison-Wesley, 1ª edition.