Introduction

Research background and motivations

MATH 

Google Scholar 
MATH 

Google Scholar 
Thomas, A., Ré, C. & Poldrack, R. Self-supervised learning of brain dynamics from broad neuroimaging data. Adv. Neural. Inf. Process. Syst. 35, 21255–21269 (2022).

Research objectives

Literature review: This study conducts a comprehensive and in-depth review of relevant domestic and international literature. The focus is on the research status, existing issues, and improvement strategies concerning the quality of public sports services. By systematically summarizing existing research findings, this study extracts valuable theoretical insights and practical guidance, laying a solid theoretical foundation and providing abundant practical reference examples.

  1. (1)

    Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

  2. (2)

    Ying Yan: Conceptualization, methodology, software, validation, formal analysis, investigation, resources, data curation, writing—original draft preparation, writing—review and editing, visualization, supervision, project administration, funding acquisition.

  3. (3)

    Technical model based on supervised learning.

  4. (4)

    Figure 1demonstrates that the innovation of the theoretical framework lies in its comprehensiveness and foresight. It innovatively integrates public sports service theory, service quality theory, and IT application theory, constructing a multidimensional and interdisciplinary analytical framework37. This innovation not only breaks through the limitations of traditional research on single theories but also fully considers the driving role of modern IT applications in optimizing the quality of public sports services. Furthermore, this theoretical framework also emphasizes practical application, highlighting the close integration of theoretical guidance and empirical research, furnishing feasible ideas and strategies for improving the quality of public sports services38. Hence, the theoretical framework’s innovativeness plays an essential guiding role in promoting research and practice in public sports services.

MATH 

Google Scholar 

Literature review

Raabe, D., Nabben, R. & Memmert, D. Graph representations for the analysis of multi-agent Spatiotemporal sports data. Appl. Intell. 53 (4), 3783–3803 (2023).Pajak, G. et al. An approach to sport activities recognition based on an inertial sensor and deep learning. Sens. Actuators A: Phys. 345, 113773 (2022).MATH 

Google Scholar 
Figure 3 depicts that this study conducts a comprehensive evaluation of the model using four different data formats. These data encompass various types of public sports service data, including user feedback, facility usage, and service quality evaluations, among others. Through processing and analyzing these data, the model demonstrates excellent performance. The evaluation results indicate that the model maintains an accuracy of over 88% for handling diverse types of data, which is quite satisfactory. Accuracy is a crucial metric for measuring whether the model’s predictions are correct, and an accuracy exceeding 88% implies that the model can accurately capture data features and make correct predictions. Moreover, the recall also exceeds 88%, further confirming the effectiveness of the model in identifying relevant data. Recall measures the model’s ability to find all relevant instances, and a high recall suggests that the model can identify as many relevant data points as possible. Thus, the model constructed in this study exhibits remarkable advantages and potential in optimizing public sports service quality. It can accurately handle and analyze various types of data, affording robust support for enhancing public sports service quality. These findings validate the feasibility and effectiveness of the model and lay a solid foundation for subsequent research and applications.

Table 1 Research status statistics.
Full size table

MATH 

Google Scholar 

Research model

This work was supported by Scientific Research Project of the Education Department of Hunan Province (Excellent Young Scholars Project): Research on the Risk Control Mechanism and Safety Assurance Strategies for College Students’ Outdoor Sports (Project Number: 24B0740).ADS 
PubMed 
PubMed Central 
MATH 

Google Scholar 
ADS 
CAS 
PubMed 
PubMed Central 
MATH 

Google Scholar 
Surek, G. A. S. et al. Video-based human activity recognition using deep learning approaches. Sensors 23 (14), 6384 (2023).

Fig. 1
figure 1
Weeraddana, N. & Premaratne, S. Unique approach for cricket match outcome prediction using Xgboost algorithms. J. Theoretical Appl. Inform. Technol. 99 (9), 2162–2173 (2021).
Full size image

CAS 
MATH 

Google Scholar 
Bai, Z. & Bai, X. Sports big data: management, analysis, applications, and challenges. Complexity, 2021: 1–11. (2021).

Fig. 2
figure 2
Fasihi, L. et al. Artificial intelligence used to diagnose osteoporosis from risk factors in clinical data and proposing sports protocols. Sci. Rep. 12 (1), 18330 (2022).
Full size image

Sharma, H. K., Choudhury, T. & Kandwal, A. Machine learning based analytical approach for geographical analysis and prediction of Boston City crime using geospatial dataset. GeoJournal, 88(Suppl 1): 15–27. (2023).College of Physical Education, Hunan City University, Yiyang, 413000, Hunan, China

  1. (1)

    Second, the supervised learning model’s application in service quality optimization is gradually becoming prominent. This model can predict future service demands, optimize resource allocation, and improve service quality by training and learning from historical data11. In the field of public sports, some studies have attempted to apply the supervised learning model to evaluate and predict service quality, achieving certain effectiveness12. However, current research is still in its early stages, and aspects such as model selection, data processing, and result interpretation require further improvement.

  2. (2)

    This article does not contain any studies with human participants or animals performed by any of the authors. All methods were performed in accordance with relevant guidelines and regulations.

  3. (3)


    Google Scholar     

  4. (4)

    Results of model basic evaluation (a: text; b: images; c: videos; d: comprehensive data).

MATH 

Google Scholar 
The influence of optimized public sports service quality on public sports participation and sports health level has been deeply explored. By comparing and analyzing the data changes before and after optimization, the actual effect of optimization measures is evaluated, and the deep reasons and mechanisms behind it are discussed, to afford a scientific basis for policy formulation.

Experimental design and performance evaluation

Datasets collection

MathSciNet 

Google Scholar 
PubMed 
MATH 

Google Scholar 

Experimental environment

As a vital part of the development of modern society, public sports service is not only related to the improvement of national physique but also an important embodiment of the prosperity and development of cultural and sports undertakings1. However, with the growing public demand for sports services, how to improve the quality of public sports services to meet diversified and personalized needs has become an urgent problem to be solved2. At the same time, the swift progress of information technology (IT), especially the extensive application of big data, artificial intelligence, and other technologies, has provided new ideas and methods for the optimization of public sports services3.

Table 2 The experimental environment design.
Full size table

Parameters setting

Evaluation of model application effects (a is text, b is image, c is video, d is comprehensive data).

Table 3 Model parameters.
Full size table

Performance evaluation

MATH 

Google Scholar 
Ying Yan

Fig. 3
figure 3
Nguyen, N. H. et al. The application of machine learning and deep learning in sport: predicting NBA players’ performance and popularity. J. Inform. Telecommunication. 6 (2), 217–235 (2022).
Full size image

The theoretical framework of this study is mainly based on the theory of service quality, public sports service, and IT applications. Firstly, the service quality theory is the cornerstone of this study. This theory emphasizes that service providers should pay attention to customers’ needs and expectations, and satisfy customers’ satisfaction by providing high-quality services. In the field of public sports services, this means the need to pay attention to the needs and expectations of the public on sports facilities, service processes, personnel quality, and so on, and optimize the service quality accordingly. Secondly, the public sports service theory provides concrete theoretical support for this study. The theory emphasizes the public welfare and universality of public sports services, aiming to promote the development of national fitness by offering high-quality sports services. Based on the relevant viewpoints of public sports service theory, this study analyzes the current situation and influencing factors of public sports service quality and proposes targeted optimization strategies21,22,23. Finally, the IT application theory furnishes theoretical support for applying a supervised learning model in optimizing public sports service quality. With the continuous development of IT, machine learning techniques such as supervised learning models have been widely used in various fields. In the field of public sports services, the supervised learning model can be used to conduct training and learning from historical data, predict future service demand, optimize resource allocation, and improve service quality24,25,26,27,28.Bunker et al. (2021) utilized the supervised learning model to successfully predict future public sports service demand by analyzing historical service data. The model effectively identified the seasonality, periodicity, and sudden changes in service demand based on facility usage, personnel flow data, and user feedback. This provided a scientific basis for the planning of public sports facilities, personnel allocation, and adjustment of service content, contributing to the improvement of service efficiency and quality13. Pelati et al. (2022) applied supervised learning algorithms to construct a multidimensional service quality evaluation model. This model comprehensively considered aspects such as service attitude, facility conditions, and service processes. By training and learning from a large amount of user evaluation data, the model achieved an objective and accurate assessment of service quality. The research results indicated that the supervised learning model exhibited high accuracy and stability in service quality evaluation, offering strong support for the continuous improvement of service quality14. Tang et al. (2021), through a comparative analysis of the application effects of different supervised learning models in optimizing public sports services, proposed a data-driven service optimization strategy. This strategy involved real-time monitoring of service data, utilizing model predictions of changing service demand trends and adjusting service content and methods accordingly. Practical results demonstrated that this data-driven service optimization strategy based on supervised learning models notably enhanced service satisfaction and user retention, thus playing a crucial role in improving the public sports services’ overall competitiveness15. The existing studies are summarized in Table 1.Parameter setting is a crucial step in constructing and training supervised learning models to optimize the quality of public sports services. Parameter settings determine the complexity and learning ability of the model and directly affect the model’s performance. The design results of model parameters are exhibited in Table 3.

Fig. 4
figure 4
Martin, R. K. et al. Unsupervised machine learning of the combined Danish and Norwegian knee ligament registers: identification of 5 distinct patient groups with differing ACL revision rates. Am. J. Sports Med. 52 (4), 881–891 (2024).
Full size image

MATH 

Google Scholar 

Discussion

Ramkumar, P. N. et al. Sports medicine and artificial intelligence: a primer. Am. J. Sports Med. 50 (4), 1166–1174 (2022).Tang, Y. et al. Triple cross-domain attention on human activity recognition using wearable sensors. IEEE Trans. Emerg. Top. Comput. Intell. 6 (5), 1167–1176 (2022).

Conclusion

Research contribution

Chen, L. & Li, S. Human motion target posture detection algorithm using semi-supervised learning in internet of things. IEEE Access. 9, 90529–90538 (2021).

Future works and research limitations

Chmait, N. & Westerbeek, H. Artificial intelligence and machine learning in sport research: an introduction for non-data scientists. Front. Sports Act. Living. 3, 363 (2021).