e-Science meets high-performance football
The Institute for Training Science and Sports Informatics receives new DFG funding.
In recent years, large and above all complex amounts of data have been generated in various fields of application related to sport. For example, in professional football in the Bundesliga and almost all other major European leagues, an incredible amount of data is collected: traditional event data such as passes, video data (e.B. sending and scouting feeds), qualitative data such as tactical patterns and attack strategies, as well as position data of the players and the ball. "These complex data sets are very heterogeneous, as accuracy and reliability vary widely," says Univ.-Prof. Dr. Daniel Memmert, Managing Director of the Institute for Training Science and Sports Informatics at the German Sports University Cologne. He and his team have been working for years on the analysis (e.B. with SOCCER©) and simulation of complex position data (so-called big data) and have developed and validated theoretical framework models for this purpose.
Now the Institute for Training Science and Sports Informatics (ITS) has been able to attract a new funding from the German Research Foundation (DFG). For the sports informatics project (ME 2678/30 1) entitled "Data-Driven Approaches for Soccer Match Analysis: an e-Science Perspective", the ITS receives more than 350,000 euros over a three-year project period. "This is an international research project with our Brazilian computer science colleagues Professor Ricardo da Silva Torres and Professor Felipe Moura," says Memmert. Together, they submitted the application to the DFG and its Brazilian counterpart FAPESP; the application is based on recent progress in the analysis of sporting events and computer science.
"The research project focuses on the analysis of complex, unstructured and multimodal data using the example of position data and video data in high-performance football," explains project employee Dr. Robert Rein, explaining the background to the DFG project. "Complex data sets require effective and efficient solutions in the areas of storage, retrieval and analysis," adds his colleague Fabian Wunderlich. The analysis covers a variety of data types such as text, audio, images, videos, graphs, and location data. The scientists are taking advantage of the scientific paradigm of e-science, "which enables large-scale calculations and communication networks to support new methods of data collection, analysis, organization, visualization and dissemination and archiving," as project leader Memmert explains.
Contact for further information:
Univ.-Prof. Dr. Daniel Memmert, Dr. Robert Rein, Fabian Wunderlich
Institut für Trainingswissenschaft und Sportinformatik
Tel.: +49 221 4982-4330