Between Dec 19, 2024 and Jan 2, 2025, datasets can be submitted to DRUM but will not be processed until after the break. Staff will not be available to answer email during this period, and will not be able to provide DOIs until after Jan 2. If you are in need of a DOI during this period, consider Dryad or OpenICPSR. Submission responses to the UDC may also be delayed during this time.
 

Evaluating the Effectiveness of Health Recommendations based on Sensor data to Support College Students' Well-being

Loading...
Thumbnail Image

Persistent link to this item

Statistics
View Statistics

Journal Title

Journal ISSN

Volume Title

Title

Evaluating the Effectiveness of Health Recommendations based on Sensor data to Support College Students' Well-being

Published Date

2024-06

Publisher

Type

Thesis or Dissertation

Abstract

Stress management is difficult for college students as college can be very demanding. Stress can contribute to serious health issues, if not timely addressed, in an individual and is a critical problem. Recent studies have highlighted the growing prevalence and severity of mental health issues among college students, exacerbated by factors such as the COVID-19 pandemic and social isolation. The interest in examining technological interventions to support stress management is on the rise. Prior work has proposed a system that generates AI based recommendations for daily stress prevention/management. The recommendation system takes into account user's preferences for activities and self-reported well-being surveys to recommend stress relievers. However, this approach is limited to one snapshot event and relies on self-report data. To address this problem, this thesis focuses on using wearable sensors to get a continuous perspective of the user's day to improve the recommendation systems' performance and user experience. We integrated wearable sensor data (i.e., Garmin activity tracker watch) within the recommendation system to track the user's daily activity levels along with the user's stress and sleep levels. We thus conducted three rounds of testing with different prioritization levels, to evaluate the recommendation system with sensor data. The evaluation study demonstrated that the model performed optimally in the second testing round with an accuracy of 94\%, validating our hypothesis by predicting the category of activities that users have not engaged in. This supports our goal of fostering diverse activity recommendations to encourage well-being. Furthermore, the user study of 20 college students revealed that the sensor-integrated system significantly improved their motivation towards self-care, with participants notably preferring the personalized, actionable health recommendations derived from real-time data. Our findings suggest that continuous monitoring via wearable sensors enhances the effectiveness of recommendation systems in stress management, offering substantial benefits in promoting healthier lifestyle choices among college students.

Keywords

Description

University of Minnesota M.S. thesis. June 2024. Major: Computer Science. Advisor: Jomara Sandbulte. 1 computer file (PDF); vii, 82 pages.

Related to

Replaces

License

Series/Report Number

Funding information

Isbn identifier

Doi identifier

Previously Published Citation

Other identifiers

Suggested citation

Putta, Mamatha. (2024). Evaluating the Effectiveness of Health Recommendations based on Sensor data to Support College Students' Well-being. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/265101.

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.