Related Experiment Video
Updated: Jul 11, 2025

Using the Threat Probability Task to Assess Anxiety and Fear During Uncertain and Certain Threat
Published on: September 12, 2014
Identifying self-disclosed anxiety on Twitter: A natural language processing approach
Daniel Zarate1, Michelle Ball2, Maria Prokofieva2
1College of Health and Biomedicine, Royal Melbourne Institute of Technology (RMIT), Australia.
Social media text analysis reveals distinct language patterns in anxiety. Machine learning accurately predicts anxiety diagnosis based on Twitter posts, highlighting negative sentiment and reduced posting frequency.
Area of Science:
- Computational linguistics
- Mental health informatics
- Social media analytics
Background:
- Social media platforms offer valuable data for mental health research.
- Text analysis of social media posts can identify patterns related to psychological conditions.
- This study focuses on understanding the digital footprint of anxiety on Twitter.
Purpose of the Study:
- To explore distinct language patterns on Twitter associated with self-reported anxiety diagnosis.
- To investigate the predictive power of language features in identifying anxiety.
- To profile user behavior and linguistic characteristics of individuals with anxiety.
Main Methods:
- Comparative analysis of 233,000 tweets from 605 users (anxious vs. non-anxious).
- Utilized natural language processing, including Linguistic Inquiry Word Count (LIWC) and sentiment analysis.
- Employed supervised machine learning models and Latent Profile Analysis (LPA).
Main Results:
- Machine learning models achieved high accuracy (up to 81.1%) in predicting anxiety diagnosis.
- Four distinct user profiles were identified: high sentiment, low sentiment, self-immersed, and normative.
- Anxious users exhibited high negative sentiment, low positive sentiment, reduced posting frequency, and longer texts.
Conclusions:
- Distinct linguistic patterns in Twitter posts correlate with self-reported anxiety.
- These patterns enable accurate prediction of anxiety diagnosis.
- Findings support the development of online mental health campaigns and awareness initiatives.
Related Concept Videos
Social Anxiety Disorder
Anxiety: Overview
Individuals with anxiety often experience a range of physical and emotional symptoms, including sweating, trembling, tachycardia, and disturbances in sleep patterns. These symptoms vary in intensity and frequency but are generally disruptive and distressing.
Anxiolytic Drugs: Overview
Primary Types of Anxiolytic Drugs
1. Benzodiazepines:
Benzodiazepines bind to the GABA-A receptor in the brain, enhancing GABA's interaction. This action reduces neurotransmission, effectively blocking anxiety-associated limbic...
Generalized Anxiety Disorder
Panic Disorder
Self-Presentation: Self-Monitoring and Self-Handicapping

