logo
ResearchBunny Logo
Machine learning of language use on Twitter reveals weak and non-specific predictions

Psychology

Machine learning of language use on Twitter reveals weak and non-specific predictions

S. W. Kelley, C. N. Mhaonaigh, et al.

Explore the groundbreaking findings of Sean W. Kelley, Caoimhe Ní Mhaonaigh, Louise Burke, Robert Whelan, and Claire M. Gillan, who investigated the potential of machine learning to predict mental health conditions from Twitter data. The study reveals intriguing insights about language patterns linked to depression, though it highlights the limitations of individualized predictions based on social media analysis.

00:00
00:00
Playback language: English
Abstract
This study investigated the ability of machine learning models to predict depression and other mental health conditions from Twitter data. Using Tweets from 1006 participants who completed questionnaires assessing symptoms of depression and 8 other mental health conditions, the researchers trained an Elastic Net model on depression severity. The model showed modest predictive performance for depression (R²=0.025), and similar or better performance in predicting other mental health problems, indicating non-specific language patterns associated with depression. The study concluded that machine learning analysis of social media data, when trained on well-validated clinical instruments, cannot make meaningful individualised predictions regarding users' mental health.
Publisher
npj Digital Medicine
Published On
Mar 25, 2022
Authors
Sean W. Kelley, Caoimhe Ní Mhaonaigh, Louise Burke, Robert Whelan, Claire M. Gillan
Tags
machine learning
depression
mental health
Twitter data
predictive performance
language patterns
clinical instruments

Related Publications

Explore these studies to deepen your understanding of the subject.

The effects of hands-on content-language integrated learning on fourth graders’ acquisition of target vocabulary and procedural knowledge in Taiwan
Education

The effects of hands-on content-language integrated learning on fourth graders’ acquisition of target vocabulary and procedural knowledge in Taiwan

C. Lai

Profiles of learners based on their cognitive and metacognitive learning strategy use: occurrence and relations with gender, intrinsic motivation, and perceived autonomy support
Education

Profiles of learners based on their cognitive and metacognitive learning strategy use: occurrence and relations with gender, intrinsic motivation, and perceived autonomy support

D. Kwarikunda, U. Schiefele, et al.

Prediction of mortality risk and duration of hospitalization of COVID-19 patients with chronic comorbidities based on machine learning algorithms
Medicine and Health

Prediction of mortality risk and duration of hospitalization of COVID-19 patients with chronic comorbidities based on machine learning algorithms

P. Amiri, M. Montazeri, et al.

Accurate and efficient molecular dynamics based on machine learning and non von Neumann architecture
Engineering and Technology

Accurate and efficient molecular dynamics based on machine learning and non von Neumann architecture

P. Mo, C. Li, et al.

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny