Detecting mental state using Machine Learning
Student
Mingqian Duan
Advisor
Ju Sun
Abstract
This project aims to use machine learning techniques to explore the relationship between patients’ mental states and their motor activity measurements. Since depression is often associated with fatigue, loss of interest in activities and sleep disturbance, I plan to investigate whether there is observable abnormalities in the motor activity level and pattern for those with major depressive disorder. The research goal is two-fold. First, meaningful and interpretable feature sets are to be extracted to describe the characteristic of the motor activity data for patients from different mental state groups. Second, an efficient machine learning algorithm will be utilized to make classifications of mental states based on motor activity patterns/levels. This will provide an objective tool to assist mental disorder diagnosis.