WCSE

Fatigue Detection Using Computer Mouse Operation Patterns

Abstract— With the spread of the Internet, the number of users utilizing digital devices has increased. Consequently, the proportion of users who use these devices for extended periods has also risen. While the widespread adoption of digital devices enhances users' daily lives, it has also introduced new issues, such as eye strain, shoulder stiffness, and wrist fatigue due to prolonged use. To address these issues, fatigue detection methods have been researched. However, many conventional methods require specialized sensors or external devices, limiting their practicality for everyday use. Therefore, this study proposes a fatigue detection method that does not require additional equipment, utilizing computer mouse operation patterns. This method focuses on the correlation coefficient between vertical mouse movement and the frequency of left clicks to determine fatigue states. Evaluation experiments showed that the proposed method achieved a higher precision value in detecting fatigue states compared to conventional machine learning models, such as Random Forest and Logistic Regression.

Index Terms— Data analysis, Fatigue detection, Machine learning, Computer mouse

Ryosuke Hagimoto, Eiji Kamioka, Chanh Minh Tran, Phan Xuan Tan
Graduate School of Engineering and Science, Shibaura Institute of Technology, JAPAN

[Download]

Cite: Ryosuke Hagimoto, Eiji Kamioka, Chanh Minh Tran, Phan Xuan Tan, "Fatigue Detection Using Computer Mouse Operation Patterns", 2025 the 15th International Workshop on Computer Science and Engineering (WCSE 2025), pp. 195-200, Jeju Island, South Korea, June 28-30, 2025.