Naslov
On the Application of Artificial Intelligence in the Problem of COVID-19 Spread Prediction

Autori
Baressi Šegota, Sandi ; Anđelić, Nikola ; Lorencin, Ivan ; Mrzljak, Vedran ; Car, Zlatan

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, prošireni sažetak, ostalo

Izvornik
5th International Workshop on Data Science - Abstract Book / Lončarić, Sven ; Šmuc, Tomislav - Zagreb : Znanstveni centar izvrsnosti za znanost o podatcima i kooperativne sustave, 2020, 35-37

Skup
5th Int'l Workshop on Data Science (IWDS 2020)

Mjesto i datum
Online, 24.11.2020

Vrsta sudjelovanja
Poster

Vrsta recenzije
Domaća recenzija

Ključne riječi
COVID-19 ; Artificial Intelligence ; Machine Learning ; Data Science ; Regression Modeling

Sažetak
Both the used methods generated high quality models, with R2 values higher than 0.98 for all the outputs. It can be concluded that AI methods can be used to achieve a good model of epidemiological curves and their components. Additionally, some differences between the algorithms can be noted. For example, while MLP is, on average, faster to train in comparison to the GP – allowing for a faster model generation ; models obtained by the MLP are not language-agnostic making them harder to implement. Models generated by GP are given in form of equations which are then easily transformed into an appropriate form to be utilized in various forms of software packages or programming languages allowing for a faster implementation of those models into new or existing tools.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo

POVEZANOST RADA