Traditional health management programs suffer from insufficient personalization and delayed real-time feedback. To address this, this paper constructs a personalized health management system based on reinforcement learning. Using a Support Vector Machine (SVM), the system collects students' health data in real time and dynamically optimizes health management strategies using a reinforcement learning algorithm. Through intelligent perception and analysis of students' health status, the system adjusts dietary and exercise recommendations in real time, providing personalized health interventions. The experiment involves university students aged 18-24 with a predisposition to obesity (BMI ≥ 24). A group experiment is conducted (the experimental group used this personalized health management system, while the control group follows a fixed exercise plan and dietary recommendations). Results show that after four weeks, the overall weight change in the experimental group ranges from 1.4 kg to 2.0 kg, and the BMI change is from 0.3 to 0.5, demonstrating the effectiveness of this program in weight loss and weight control.

Jian Li, Zhou Zhou. Intelligent Health Management Solutions for Obese College Students in and out of Class Based on AI Technology. The Frontiers of Society, Science and Technology (2025), Vol. 7, Issue 7: 49-55. https://doi.org/10.25236/FSST.2025.070708.

[1] Adirahma, A. S., Widiyanto, W. E., Setyawan, H., et al. (2024). Badminton training management strategy with circuit training method to improve athlete performance among students. Retos: Nuevas tendencias en educación física, deporte y recreación, 61, 108-120.

[2] Hong, Y., Ullah, R., Wang, J. B., et al. (2023). Trends of obesity and overweight among children and adolescents in China. World Journal of Pediatrics, 19(12), 1115-1126.

[3] Johnson, S., Gupta, S., Mackey, E., et al. (2023). “We feel like we are in it alone”: A mixed-methods study of pediatric primary care barriers for weight management. Childhood Obesity, 19(1), 3-12.

[4] Barlas, T., Altinova, A. E., Akturk, M., et al. (2024). Credibility of ChatGPT in the assessment of obesity in type 2 diabetes according to the guidelines. International Journal of Obesity, 48(2), 271-275.

[5] Lofton, H., Ard, J. D., Hunt, R. R., et al. (2023). Obesity among African American people in the United States: A review. Obesity, 31(2), 306-315.

[6] Ahmed, M. M., Ibrahim, R. M., & Younis, N. M. (2024). Assessment of nursing students’ attitudes toward obese individuals. Journal of Current Medical Research and Opinion, 7(05), 2468-2476.

[7] Woldemariam, S., Dorner, T. E., Wiesinger, T., et al. (2023). Multi-omics approaches for precision obesity management: Potentials and limitations of omics in precision prevention, treatment, and risk reduction of obesity. Wiener Klinische Wochenschrift, 135(5), 113-124.

[8] Salminen, P., Kow, L., Aminian, A., et al. (2024). IFSO consensus on definitions and clinical practice guidelines for obesity management—An international Delphi study. Obesity Surgery, 34(1), 30-42.

[9] Dayyeh, B. K. A., Stier, C., Alqahtani, A., et al. (2024). IFSO bariatric endoscopy committee evidence-based review and position statement on endoscopic sleeve gastroplasty for obesity management. Obesity Surgery, 34(12), 4318-4348.

[10] Okafor, M. T. (2024). Personalized nutrition for obesity management: A mini-review. International Journal of Medicine and Health Development, 29(4), 285-288.

[11] Elmaleh-Sachs, A., Schwartz, J. L., Bramante, C. T., et al. (2023). Obesity management in adults: A review. Jama, 330(20), 2000-2015.