Faculty Advisor

Agu, Emmanuel O.

Faculty Advisor

Rundensteiner, Elke A.


The hallmark indicator of depressive disorders is a presence of “sad, empty, or irritable mood, accompanied by somatic and cognitive changes that significantly affect the individual’s capacity to function”. The overall goal of our project is to provide a tool for doctors to effortlessly detect depression, and in effect achieve greater coverage in detecting depression over the general population. We use machine learning techniques to create a mobile application that infers a smartphone user’s severity of depression from data scraped off their phone and social media websites. Through our study, we have demonstrated the feasibility of this approach to diagnosing depression, achieving an average testset RMSE of 5.67 across all modalities in the task of PHQ-9 score predictions.


Worcester Polytechnic Institute

Date Accepted

March 2018


Computer Science

Project Type

Major Qualifying Project



Advisor Department

Computer Science