TL;DR
This paper presents an intelligent sports training system for college students, leveraging deep learning and AI to address declining physical fitness and inefficient venue management. The system offers personalized sports program and course recommendations, aiming to improve students' physical and mental well-being.
What Problem Does This Paper Address?
College students today face a dual challenge: declining physical fitness and increasing mental health concerns, including anxiety and depression. This is exacerbated by inefficient management of college sports venues, leading to underutilization of resources and limited access for students. Traditional approaches to sports education and venue management are struggling to keep pace with the growing enrollment and the need for personalized, effective health interventions. The authors aim to bridge this gap by developing an AI-powered system that not only optimizes venue management but also promotes physical activity and mental health through intelligent recommendations.
Methods
The core of the system is an intelligent sports management system built using a Browser/Server (B/S) architecture and a distributed microservice system based on the Spring Cloud framework. This architecture allows for modularity and scalability.
Key methodological components include:
- Deep Learning and Artificial Intelligence: The system utilizes deep learning for its recommendation algorithms. This involves analyzing user data such as age, BMI, and physical health status to suggest suitable sports programs.
- Recommendation Algorithms:
- Content-based filtering is employed to analyze user preferences and item attributes (e.g., fitness equipment features).
- Collaborative filtering is also used to find users with similar interests and predict preferences.
- Specific similarity calculation methods like cosine similarity, co-occurrence similarity, and Pearson correlation coefficient are discussed.
- A hybrid intelligent sports scheme recommendation model is designed, combining case reasoning with sports health knowledge ontology.
- Microservice Architecture: Spring Cloud is used to integrate various components into a distributed microservice system, enhancing development efficiency and maintainability.
- Database Design: An Entity-Relationship (E-R) diagram is used to model the database structure, focusing on venue and course information.
- User Interface: A B/S structure is chosen for its ease of development and maintenance, with a presentation layer for user interaction.
Datasets
The paper mentions the use of data from Shanghai X University for a questionnaire survey. The survey involved 496 valid questionnaires from college students, utilizing:
- Physical Activity Rating Scale-3 (PARS-3): To assess the amount of physical exercise.
- Self-Rating Depression Scale (SDS): To measure depression levels.
- Self-Rating Anxiety Scale (SAS): To measure anxiety levels.
- Symptom Checklist 90 (SCL-90): A mental health scale covering various symptom dimensions.
No new benchmarks were introduced; rather, existing scales were used for analysis.
Key Insights and Ideas
- AI for Holistic Student Well-being: The system demonstrates how AI can move beyond simple task automation to address complex issues like student physical fitness and mental health simultaneously.
- Personalized Recommendations are Key: The success of the system hinges on its ability to provide tailored sports programs and course recommendations, moving away from one-size-fits-all approaches.
- Microservices for Scalability and Agility: The adoption of a microservice architecture with Spring Cloud is crucial for building a robust and adaptable system that can evolve with user needs and data.
- Bridging the Gap Between Physical Activity and Mental Health: The research highlights a significant correlation between exercise levels and reduced anxiety/depression scores, underscoring the importance of promoting sports lifestyles.
Key Results
- Functional Test: The system's functional tests showed that it could run well, accurately displaying venue information, completing online bookings, and demonstrating compatibility across multiple systems.
- Questionnaire Survey Analysis:
- 64% of students increased their weekly exercise frequency after the system's promotion.
- Significant differences were found in anxiety and depression scores based on exercise amount.
- Groups with medium and large amounts of exercise showed significantly lower anxiety scores (p < 0.001) and depression scores (p < 0.001 for medium, p = 0.003 for large compared to medium).
- The larger the amount of exercise, the lower the scores for anxiety and depression.
Limitations and Future Work
- System Optimization: The authors acknowledge that the system, while functional, requires continuous optimization in terms of function, performance, operability, compatibility, and security.
- Big Data and Data Mining: With the accumulation of system data, they suggest the need to incorporate big data and data mining technologies to establish a data warehouse for statistical analysis and support school sports management decision-making.
- Data Availability: The datasets used in the study are restricted by university regulations and are not readily available. Requests should be directed to the corresponding author.
Summary updated