diplomsko delo
Vid Rebol (Author), Igor Kononenko (Mentor), Petar Vračar (Co-mentor)

Abstract

Na raziskovalnem področju košarke je bilo predstavljenih že veliko metod za simulacijo in napovedovanje zmagovalca tekem. Cilj realistične simulacije je skladnost generirane dinamike statistike z dinamiko statistike iz dejanskih tekem in posledično čim bolj natančna napoved zmagovalca tekme. Naš pristop opisuje statistiko ekip v različnih časovno-rezultatskih situacijah. Raziskujemo prenos analize števcev v teh situacijah in njihovo uporabnost pri simulaciji. V prvem delu razvijemo modele, ki aproksimirajo obnašanje ekip. V drugem delu povežemo dobljene aproksimacije z modelom za napovedovanje naslednjega dogodka na košarkarski tekmi. S pomočjo tega modela tvorimo simulacije, ki jih primerjamo s potekom dejanskih tekem. Pristop situacijskega modeliranja igralnega sloga se v splošnem ni izkazal za uspešnega pri generiranju realističnih simulacij.

Keywords

igralni slog;nevronska mreža;situacijsko modeliranje;simuliranje tekem;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [V. Rebol]
UDC: 004.94(043.2)
COBISS: 142199299 Link will open in a new window
Views: 32
Downloads: 9
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Playing style modelling of basketball games
Secondary abstract: Many simulation and winner prediction methods were already proposed in the research area of basketball. A goal of a realistic simulation is the compliance of generated statistics dynamics with the statistic dynamics from real games. Our approach describes team statistics in different time-result situations. We research the transfer of statistics analysis corresponding to a situation and their usefulness in simulation. In the first part we develop models that approximate the behaviour of teams. In the second part we connect the obtained approximations with the model that predicts the next event in a basketball game. This model helps us create simulations, which we compare with the actual course of basketball games. The situational playing style modelling approach has generally not proven to be successful in generating realistic simulations.
Secondary keywords: playing style;neural network;situational modelling;game simulation;computer science;diploma;Računalniška simulacija;Košarka;Modeliranje podatkov (računalništvo);Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 49 str.
ID: 18003275