Naslov
Prediction of Euroleague Games based on Supervised Classification Algorithm k-Nearest Neighbours

Autori
Horvat, Tomislav ; Job, Josip ; Medved, Vladimir

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the 6th International Congress on Sport Sciences Research and Technology Support / Pezarat-Correia, Pedro ; Vilas-Boas, João ; Rivera, Octavio ; Cabri, Jan - Sevilla : SCITEPRESS, 2018, 203-207

ISBN
978-989-758-325-4

Skup
6th International Congress on Sport Sciences Research and Technology Support (icSPORTS 2018)

Mjesto i datum
Sevilla, Španjolska, 20.09.2018. - 21.09.2018

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Basketball, Classification, Database Analysis, Euroleague, Information System, k-Nearest Neighbours, Machine Learning, Prediction

Sažetak
Machine learning becomes one of the top fields in world of ICT (Information and communications technology). In the last few decades machine learning has taken a big boost in sports. Sport fans are using machine learning in sport betting and sport managers are using it for player selection, performance evaluation and even in outcome prediction. This paper gives basketball outcome prediction algorithms using k-nearest neighbours (k-nn). Few methods of preparing data calculated with different k and time period for k-nn algorithm will be presented and closely explained. Feature selection will also be discussed and results based on feature selection will be presented.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Informacijske i komunikacijske znanosti

POVEZANOST RADA