Depression is a serious mental health problem affecting the mind and body of people in our society. As of today, depression affects about 5 to 7% of the American population, which roughly translates to 14 million people! A particularly worrying concern these days is the increasing incidence of depression among college students. The problem is so severe that many US campuses have dedicated centers with mental health professionals catering to students needs. Although there are very effective treatments for depression, studies report that ~80% of depressed students do not seek any help due to the lack of perception of the problem, low self esteem, and loneliness.
The aim of this thesis is to develop a transparent depression identification/ classification framework that can be deployed in college settings. The premise of this thesis stems from surveys in the mental health community revealing extensive known correlations between Internet use and Depression. The thesis investigates the feasibility of identifying and classifying depression via mining real Internet usage patterns derived from Cisco Netflow data.
To the best of the authors knowledge, this thesis is the first to study depression using real Internet data. As a result, several new statistical results correlating Internet use with depression is presented. Additionally, this thesis takes a quantum leap forward in transparent depression detection by developing a classifier that can predict depression with ~74% accuracy, demonstrating the feasibility of the approach. The performance was further improved to ~84% by considering deviations in baseline Internet activity of a user.
The strength of the proposed framework lies in its practicality, extensibility, transparency, and privacy preserving nature"--Abstract, page iii.
Wunsch, Donald C.
Madria, Sanjay Kumar
M.S. in Computer Science
Missouri University of Science and Technology
ix, 92 pages
© 2011 Raghavendra Kotikalapudi, All rights reserved.
Thesis - Restricted Access
Depression, Mental -- Diagnosis
College students -- Psychology
Internet users -- Statistics -- Data processing.
Print OCLC #
Electronic OCLC #
Link to Catalog Record
Electronic access to the full-text of this document is restricted to Missouri S&T users. Otherwise, request this publication directly from Missouri S&T Library or contact your local library.http://merlin.lib.umsystem.edu/record=b10737199~S5
Kotikalapudi, Raghavendra, "A framework for transparent depression classification in college settings via mining internet usage patterns" (2011). Masters Theses. 7360.
Share My Thesis If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.