A synthetic data-driven machine learning approach for athlete performance attenuation prediction

Athlete performance monitoring is effective for optimizing training strategies and preventing injuries. However, applying machine learning (ML) frameworks to this domain remains challenging due to data scarcity limitations. This study extends previous research by evaluating Tabular Variational Autoe...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Frontiers in sports and active living Ročník 7; s. 1607600
Hlavní autoři: Cordeiro, Mauricio C., Cathain, Ciaran O., Daly, Lorcan, Kelly, David T., Rodrigues, Thiago B.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Switzerland Frontiers Media S.A 27.05.2025
Témata:
athlete monitoring
machine learning
performance prediction
Sports and Active Living
synthetic data
tabular variational autoencoders
performance prediction
athlete monitoring
tabular variational autoencoders
machine learning
synthetic data
ISSN:2624-9367, 2624-9367
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Athlete performance monitoring is effective for optimizing training strategies and preventing injuries. However, applying machine learning (ML) frameworks to this domain remains challenging due to data scarcity limitations. This study extends previous research by evaluating Tabular Variational Autoencoders (TVAE) for generating synthetic data to predict performance attenuation in Gaelic football athletes. This study assesses synthetic data quality through a comprehensive evaluation framework combining column shape similarity metrics and Hellinger distance analysis, quantifying distributional fidelity across individual variables. Our ML implementation follows a two-phase approach. In the first phase, we evaluated models trained on hybrid datasets with varying synthetic proportions (10%-100%). In the second phase, we examined models trained exclusively on synthetic data and tested them on real data to analyze the utility of the synthetic data. Our results demonstrate that TVAE-generated synthetic data closely replicates original distribution patterns, achieving 85.53% column shape similarity and a Hellinger distance of 0.169. Models trained with additional synthetic data or exclusively on synthetic data outperformed real-data baselines across multiple metrics, particularly for neuromuscular parameters. Our findings emphasize that this approach increased data availability and improved model performance in specific scenarios. Several limitations remain: (1) there is limited framework transferability to sports with different physiological demands; (2) the Synthetic Data Generation (SDG) does not currently enforce feature constraints, and future implementations must ensure the procedure respects domain-specific feature limits; and (3) TVAE faced data fidelity challenges with certain variables, such as VO . These findings demonstrate the utility of synthetic data for predicting performance attenuation in Gaelic Football athletes. They address the challenge of data scarcity and highlight how synthetic data can be effectively integrated across physiological, neuromuscular, and perceptual metrics in athlete monitoring. This opens new possibilities for exploring similar classification tasks in sports performance analysis.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Soukaina Hattabi, University of Jendouba, Tunisia
Present Address: Ciaran O. Cathain,Department of Sport Science and Nutrition, Faculty of Science and Engineering, Maynooth University, Maynooth, Ireland
Reviewed by: António Miguel Monteiro, Instituto Politécnico de Bragança, Portugal
Edited by: Pedro Forte, Higher Institute of Educational Sciences of the Douro, Portugal
ISSN:2624-9367
2624-9367
DOI:10.3389/fspor.2025.1607600