Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: JMIR Public Health and Surveillance

Date Submitted: Jun 5, 2019
Date Accepted: Jan 27, 2020

The final, peer-reviewed published version of this preprint can be found here:

Classification of Health-Related Social Media Posts: Evaluation of Post Content–Classifier Models and Analysis of User Demographics

Rivas R, Sadah SA, Guo Y, Hristidis V

Classification of Health-Related Social Media Posts: Evaluation of Post Content–Classifier Models and Analysis of User Demographics

JMIR Public Health Surveill 2020;6(2):e14952

DOI: 10.2196/14952

PMID: 32234706

PMCID: 7160708

Classification of Health-Related Social Media Posts: Analysis and Evaluation

  • Ryan Rivas; 
  • Shouq A Sadah; 
  • Yuhang Guo; 
  • Vagelis Hristidis

ABSTRACT

Background:

The rising volume of health-related social media activity, where users connect, collaborate, and engage, has increased the significance of analyzing how people use them.

Objective:

The aim of this study is to classify the intent–e.g., share experiences, seek support–of users who participate in health-related social media, and study the effect of the user demographics to the posting intent.

Methods:

We analyzed two different types of health-related social media: (1) health Web forums WebMD and DailyStrength, and (2) general Web-based social networks Twitter and Google+. We identified several post intents and built classifiers to automatically detect the intent of posts. These classifiers were used to study the distribution of intents for various demographic groups.

Results:

The results of this study are: (1) we achieved accuracy of at least 84% and balanced accuracy of at least 81.25% for half the intents in our experiments; (2) the majority (70.04%) of posts by male WebMD users ask for advice; (3) male users’ WebMD posts are more likely to ask for medical advice than female users’ posts; (4) the majority (> 80%) of posts on DailyStrength share experiences, regardless of the gender, age group, or location of their authors; (5) health-related posts on Twitter are used to share experiences less frequently than posts on WebMD and DailyStrength; and (6) health-related educational material is shared on Google+ most frequently by Asian users and least frequently by users with age 35-44.

Conclusions:

We studied and analyzed the intent of users participating in health-related social media. Our results can guide health care providers and practitioners to create effective and targeted health care campaigns.


 Citation

Please cite as:

Rivas R, Sadah SA, Guo Y, Hristidis V

Classification of Health-Related Social Media Posts: Evaluation of Post Content–Classifier Models and Analysis of User Demographics

JMIR Public Health Surveill 2020;6(2):e14952

DOI: 10.2196/14952

PMID: 32234706

PMCID: 7160708

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.