Mental Health Analysis via Social Media Data

Amir Hossein Yazdavar; Mohammad Saied Mahdavinejad; Goonmeet Bajaj; Krishnaprasad Thirunarayan; Jyotishman Pathak; Amit Sheth

doi:10.1109/ICHI.2018.00102

Mental Health Analysis via Social Media Data

Amir Hossein Yazdavar, Mohammad Saied Mahdavinejad, Goonmeet Bajaj, Krishnaprasad Thirunarayan, Jyotishman Pathak, Amit Sheth

Wright State University

Research output: Contribution to journal › Article › peer-review

Abstract

With ubiquity of social media platforms, millions of people are routinely sharing their moods, feelings and even their daily struggles with mental health issues by expressing it verbally or indirectly through images they post. In this study, we aim to examine exploitation of big multi-modal social media data for studying depressive behavior and its population trend across the U.S. to better understand a regions influence on the prevailing environment and available care. In partic-ular, employing statistical techniques along with the fusion of heterogeneous features gleaned from different modalities (shared images and textual content), we build models to detect depressed individuals and their demographics.

Original language	American English
Journal	Proceedings - 2018 IEEE International Conference on Healthcare Informatics, ICHI 2018
DOIs	https://doi.org/10.1109/ICHI.2018.00102
State	Published - Jul 24 2018

Keywords

Machine Learning, Mental Health, Multi-modal Analysis, Natural Language Processing, Regression, Social Media, Statistical analysis

Disciplines

Computer Sciences
Engineering

Access to Document

10.1109/ICHI.2018.00102License: Unspecified

Cite this

@article{7e899d19dde2433f9d552ff1db03767d,

title = "Mental Health Analysis via Social Media Data",

abstract = " With ubiquity of social media platforms, millions of people are routinely sharing their moods, feelings and even their daily struggles with mental health issues by expressing it verbally or indirectly through images they post. In this study, we aim to examine exploitation of big multi-modal social media data for studying depressive behavior and its population trend across the U.S. to better understand a regions influence on the prevailing environment and available care. In partic-ular, employing statistical techniques along with the fusion of heterogeneous features gleaned from different modalities (shared images and textual content), we build models to detect depressed individuals and their demographics.",

keywords = "Machine Learning, Mental Health, Multi-modal Analysis, Natural Language Processing, Regression, Social Media, Statistical analysis",

author = "Yazdavar, {Amir Hossein} and Mahdavinejad, {Mohammad Saied} and Goonmeet Bajaj and Krishnaprasad Thirunarayan and Jyotishman Pathak and Amit Sheth",

year = "2018",

month = jul,

day = "24",

doi = "10.1109/ICHI.2018.00102",

language = "American English",

journal = "Proceedings - 2018 IEEE International Conference on Healthcare Informatics, ICHI 2018",

}

TY - JOUR

T1 - Mental Health Analysis via Social Media Data

AU - Yazdavar, Amir Hossein

AU - Mahdavinejad, Mohammad Saied

AU - Bajaj, Goonmeet

AU - Thirunarayan, Krishnaprasad

AU - Pathak, Jyotishman

AU - Sheth, Amit

PY - 2018/7/24

Y1 - 2018/7/24

N2 - With ubiquity of social media platforms, millions of people are routinely sharing their moods, feelings and even their daily struggles with mental health issues by expressing it verbally or indirectly through images they post. In this study, we aim to examine exploitation of big multi-modal social media data for studying depressive behavior and its population trend across the U.S. to better understand a regions influence on the prevailing environment and available care. In partic-ular, employing statistical techniques along with the fusion of heterogeneous features gleaned from different modalities (shared images and textual content), we build models to detect depressed individuals and their demographics.

AB - With ubiquity of social media platforms, millions of people are routinely sharing their moods, feelings and even their daily struggles with mental health issues by expressing it verbally or indirectly through images they post. In this study, we aim to examine exploitation of big multi-modal social media data for studying depressive behavior and its population trend across the U.S. to better understand a regions influence on the prevailing environment and available care. In partic-ular, employing statistical techniques along with the fusion of heterogeneous features gleaned from different modalities (shared images and textual content), we build models to detect depressed individuals and their demographics.

KW - Machine Learning, Mental Health, Multi-modal Analysis, Natural Language Processing, Regression, Social Media, Statistical analysis

UR - https://corescholar.libraries.wright.edu/cse/560

U2 - 10.1109/ICHI.2018.00102

DO - 10.1109/ICHI.2018.00102

M3 - Article

JO - Proceedings - 2018 IEEE International Conference on Healthcare Informatics, ICHI 2018

JF - Proceedings - 2018 IEEE International Conference on Healthcare Informatics, ICHI 2018

ER -

Mental Health Analysis via Social Media Data

Abstract

Keywords

Disciplines

Access to Document

Other files and links

Cite this