Twitter-STMHD: An Extensive User-Level Database of Multiple Mental Health Disorders

Abstract

Social Media is equipped with the ability to track and quantify user behavior, establishing it as an appropriate resource for mental health studies. However, previous efforts in the area have been limited by the lack of data and contextually relevant information. There is a need for large-scale, well-labeled mental health datasets with fast reproducible methods to facilitate their heuristic growth. In this paper, we cater to this need by building the Twitter - Self-Reported Temporally-Contextual Mental Health Diagnosis Dataset (Twitter-STMHD), a large scale, user-level dataset grouped into 8 disorder categories and a companion class of control users. The dataset is 60% hand-annotated, which lead to the creation of high-precision self-reported diagnosis report patterns, used for the construction of the rest of the dataset. The dataset, instead of being a corpus of tweets, is a collection of user-profiles of those suffering from mental health disorders to provide a holistic view of the problem statement. By leveraging temporal information, the data for a given profile in the dataset has been collected for disease prevalence periods: onset of disorder, diagnosis and progression, along with a fourth period: COVID-19. This is the only and the largest dataset that captures the tweeting activity of users suffering from mental health disorders during the COVID-19 period.

Vanishing Boundaries: A Unifying Account of Multidimensional Emotion Dynamics and Alterations in Depression

“…In the current work, two additional datasets were used: Twitter-STMHD [55] and HappyDB [26]. The purpose of the Twitter-STMHD dataset is to be used alongside HappyDB to train a model capable of detecting posts expressing happy moments.…”

“…The Twitter-Depression dataset is kept intact for the final analyses. Twitter-STMHD [55] contains posts from individuals with eight mental disorders (e.g., depression, anxiety, PTSD, etc. ), labeled by their mentions of diagnosis, similar to Twitter-Depression.…”

Language and Mental Health: Measures of Emotion Dynamics from Text as Linguistic Biosocial Markers

“…We use a recently compiled dataset-Twitter-STMHD (Suhavi et al, 2022). It comprises of tweets from 27,003 users who have self-reported as having a mental health diagnosis on Twitter.…”

“…A Twitter-STMHD Dataset Suhavi et al (2022) created a regular expression pattern to identify posts which contained a selfdisclosure of a diagnosis and the diagnosis name (using a lexicon of common synonyms, abbreviations, etc.) such as 'diagnosed with X'.…”

“…X (formerly Twitter) provides an abundant amount of textual data from the public. 1 By considering tweets from users who have chosen to self-disclosed as having an MHC (Suhavi et al, 2022), we can analyse the differences in UED metrics across a diagnosis and the control population. We describe how utterance emotion dynamics (UED) metrics compare between different each of the seven diagnoses (ADHD, bipolar, depression, MDD, OCD, PPD, PTSD) and a control group; and for each MHC we explore the following research questions:…”