Analysis of process dynamics in soccer by means of artificial neural networks and Voronoi-cells
Deutscher übersetzter Titel: | Eine Analyse von Prozessdynamiken im Fußball mittels künstlicher neuronaler Netze und Voronoi-Zellen |
---|---|
Autor: | Perl, Jürgen; Memmert, Daniel |
Erschienen in: | Sportinformatik X : Jahrestagung der dvs-Sektion Sportinformatik vom 10.-12. September 2014 in Wien |
Veröffentlicht: | Hamburg: Feldhaus, Edition Czwalina (Verlag), 2015, S. 130-135, Lit. |
Herausgeber: | Deutsche Vereinigung für Sportwissenschaft / Sektion Sportinformatik |
Format: | Literatur (SPOLIT) |
Publikationstyp: | Sammelwerksbeitrag |
Medienart: | Gedruckte Ressource |
Sprache: | Englisch |
Schlagworte: | |
Online Zugang: | |
Erfassungsnummer: | PU201511008209 |
Quelle: | BISp |
Einleitung
Coaches in soccer normally read and understand the complex dynamics of a game just by watching and experiencing – but then use simple code numbers like numbers of passes to classify such complex behavior. Obviously, it is difficult to bridge that gap by means of technology, modeling and analysis. On the one hand, automatic recording of the players’ position data is a first and important step in order to get basic information – but fails in registering playing processes or types and success of tactical interactions. On the other hand, reducing the information which is contained in those position data to simple statistical features does not make much sense – but is difficult to replace by both convincing and transparent information. In a joint project, funded over four years by the German Research Foundation (DFG), some novel modelling approaches and objective analysis tools have been developed, which can help to better understand and evaluate tactical behavior in games. Einleitung