Automated Detection of Match Phases in Football from Spatio-Temporal Tracking Data Using Graph Neural Networks

Key points

• This paper proposes a framework combining graph neural networks (GNNs) with a sequential model to classify match phases on a second-by-second basis across a seven-class taxonomy.1
• Match phase classification is tactically meaningful, and the availability of rule-based labels across 203 matches makes it a suitable testbed for a systematic comparison of adjacency constructions and message-passing layers, a question that has received limited attention in existing research.1
• Graph representations using a domain-informed Delaunay triangulation that approximates passing lanes, paired with a custom Spatial Edge-Augmented Convolution (SEAConv) layer that injects edge attributes directly into messages, achieve the best performance by capturing spatial dependencies while limiting uninformative messages from redundant edges.1
• Integrated Gradients attributions indicate the model uses spatial player configurations, particularly horizontal positioning, in combination with possession and ball-status indicators.1

Sources (1)

• [1]Automated Detection of Match Phases in Football from Spatio-Temporal Tracking Data Using Graph Neural Networks
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 09:27 AM

  Spatio-temporal tracking data has opened new possibilities for detecting complex tactical patterns in football, yet modeling the interactive movements of multiple players remains challenging.

  This paper proposes a framework combining graph neural networks (GNNs) with a sequential model to classify match phases on a second-by-second basis across a seven-class taxonomy.

Extractive summary: sentences quoted from the sources.