It is very important to identify mental health problems early and efficiently, but traditional method relies on face-to-face communication which suffers from the limitations in practice. This study aimed to propose an innovative method of detecting mental health problems via web use behaviors. 102 graduates were administrated by SCL-90 questionnaire to get their actual mental health status with 10 dimensions, and their web use behaviors were acquired from Internet access log recorded on the gateway. A computational model for predicting scores on each SCL-90 dimension was built based on web use behaviors. Results indicated that the value of Pearson Correlation Coefficient between predicted scores and actual scores on each dimension ranged from 0.49 to 0.65, and the value of Relative Absolute Error (RAE) ranged from 75% to 89%. It suggests that it is efficient and valid to identify mental health status through web use behaviors, which would improve the performance of mental health care services in the future. © Springer International Publishing 2013.
CITATION STYLE
Li, A., Zhang, F., & Zhu, T. (2013). Web use behaviors for identifying mental health status. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8211 LNAI, pp. 348–358). https://doi.org/10.1007/978-3-319-02753-1_35
Mendeley helps you to discover research relevant for your work.