Show simple item record

dc.contributor.authorKim, Seunghyun
dc.identifier.otherbibid: 11050269
dc.description.abstractDepression has been an ongoing mental health issue that has been affecting a wide range of humanity, particularly the young adults. To address and observe the more general public in a natural habitat, social media is examined for constructing a system to accurately detect depression. Despite the assiduous effort to construct a novel mechanism to detect depression from social media, behavioral approaches had underlying problems for users with a short activity span. To address this problem, emotion analysis was used as a tool to extract the emotion(s) of a user’s post to identify those with depression. Via machine learning techniques to construct an emotion classifier which in turn creates emotion embeddings for a binary classifier, this study proposes a pipeline structure to identify reddit posts from the depression subreddit. The model yielded promising results, introducing emotional analysis as a novel methodology in assessing mental health within social media.
dc.subjectComputer science
dc.subjectSocial Media
dc.subjectEmotional Analysis
dc.titleDetecting Depression in Social Media : An Emotional Analysis Approach
dc.typedissertation or thesis Science University of Science, Computer Science
dc.contributor.chairCardie, Claire T.
dc.contributor.committeeMemberBunea, Florentina

Files in this item


This item appears in the following Collection(s)

Show simple item record