Towards personalised performance prediction in road cycling through machine learning
We study the feasibility of applying machine learning to predict the performance of road cyclists using publicly available data. The performance is investigated by predicting the presence or absence in the top places of next year`s ranking based on a rider`s characteristics and results in the current and previous years. We apply several classification algorithms and obtain that random forest is the best-performing model. Our work is a first step towards creating personalised performance models in professional road cycling.
© Copyright 2023 13th World Congress of Performance Analysis of Sport and 13th International Symposium on Computer Science in Sport. IACSS&ISPAS 2022. Published by Springer. All rights reserved.
| Subjects: | |
|---|---|
| Notations: | endurance sports technical and natural sciences |
| Tagging: | maschinelles Lernen Algorithmus |
| Published in: | 13th World Congress of Performance Analysis of Sport and 13th International Symposium on Computer Science in Sport. IACSS&ISPAS 2022 |
| Language: | English |
| Published: |
Cham
Springer
2023
|
| Online Access: | https://doi.org/10.1007/978-3-031-31772-9_20 |
| Volume: | 1448 |
| Pages: | 93-96 |
| Document types: | article |
| Level: | advanced |