Name
Abstract | ||
|---|---|---|
In professional football, every play data is recorded such as Pass, Dribble, etc. However, the play data is manually recorded, which requires huge effort. To reduce the human effort, we propose a method to recognize the labels of plays in football games from tracking data. By using features extracted from tracking data, we generate a play classifier model based on machine learning. We have evaluated the proposed method through real tracking data recorded in Japan Professional Football League (J. League). The results have shown that our play recognition is effective for mitigating the heavy workload for play labeling. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1007/978-3-319-98530-5_78 | ADVANCES IN NETWORK-BASED INFORMATION SYSTEMS, NBIS-2018 |
Field | DocType | Volume |
Football,Workload,Computer science,League,Tracking data,Artificial intelligence,Classifier (linguistics),Machine learning | Conference | 22 |
ISSN | Citations | PageRank |
2367-4512 | 0 | 0.34 |
References | Authors | |
5 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Tomoki Imai | 1 | 0 | 0.34 |
| Akira Uchiyama | 2 | 78 | 14.48 |
| Takuya Magome | 3 | 0 | 0.34 |
| Higashino, T. | 4 | 19 | 15.19 |