A Two-Level Multi-Modal Analysis for Depression Detection From Online Social Media

A Two-Level Multi-Modal Analysis for Depression Detection From Online Social Media

Dhrubasish Sarkar, Piyush Kumar, Poulomi Samanta, Suchandra Dutta, Moumita Chatterjee
Copyright: © 2022 |Pages: 22
DOI: 10.4018/IJSI.309114
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

According to World Health Organization statistics, depression is a prominent cause of concern worldwide, leading to suicide in the majority of these cases if left untreated. Nowadays, social media is a great place for users to express themselves through text, emoticons, images, etc., which reflect their thoughts and moods. This has opened up the possibility of studying social networks in order to better comprehend the mental states of their participants. The primary goal of the research is to examine Twitter user personas and tweets in order to uncover traits that may signal depressive symptoms among online users. A two-level depression detection method is proposed in which suspected depressed individuals are identified using social media features, personality traits, temporal and sentiment analysis of user biographies. Using the support vector machine classifier, these qualities are integrated with additional linguistic and topic features to achieve an accuracy of 89%. According to the research, effective feature selection and their combinations aid in enhancing performance.
Article Preview
Top

1. Introduction

Mental illness is a major concern in terms of public health issue worldwide because of the pain, morbidity, dysfunction, and the economic burden that it brings. In National Mental Health Survey 2015-16, World Health Organization (WHO) reported an estimate of roughly 15% adults in India require professional care for more than one mental health issue (“Depression”, 2022). While there are many classes of mental disorders in India, one of the most common of them is depression (“India - share of mental disorders among adults by classification 2017 | Statista”, 2022). According to Lancet studies, there is a significant correlation between depressive illnesses and the suicide death rate at the state level in India (Sagar et al., 2020). According to World Health Organization’s Comprehensive Mental Health Action Plan, depression is one of the most common mental diseases worldwide impacting more than three hundred million people globally. Depression is mostly characterized by persistent sadness and lack of interest in previously enjoyed activities as well as an inability to carry out daily activities for at least two weeks (Zimmerman & Coryell, 1987). An early diagnosis of depression is crucial for effective treatment (Ríssola, Aliannejadi & Crestani, 2020). However, because of the social stigma associated with depressive symptoms, a large proportion of persons suffering from depressive symptoms continue to avoid obtaining professional care (Zou, Li & Cho, 2020). As a result, they often resort to informal sources like social media to solve their problems.

Social networking sites (such as Twitter, and Facebook) have grown common for people coming together and sharing their experiences and feelings. Online forums, tweets and blogs provided a way of relief for people suffering from mental health issues as they could implicitly or explicitly share their feelings and experiences (Park et al., 2012; Bathina et al., 2021). General description of a social media user contains a profile and a set of posts. The profile comprises of features such as name, age, and location which describes the user’s persona. Posts, on the other hands are contents that a user shares and can contain text, photograph, video, or audios. Social media allows for the analysis of social network data for user moods and emotions with the goal of studying their behavior while on social media. When depressed users use social media, they often tend to behave differently, which results in a large amount of data that could be used to model various characteristics. Psychological research has established a substantial relation between an individual’s emotional well-being and their language use. Using this idea, many researchers have developed new types of potential health care solutions and approaches for detecting early stages of depression. This is achieved by utilizing Machine learning algorithms along with Natural Language Processing (NLP) techniques for identifying depression in user posts.

Complete Article List

Search this Journal:
Reset
Volume 12: 1 Issue (2024)
Volume 11: 1 Issue (2023)
Volume 10: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 9: 4 Issues (2021)
Volume 8: 4 Issues (2020)
Volume 7: 4 Issues (2019)
Volume 6: 4 Issues (2018)
Volume 5: 4 Issues (2017)
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing