Related Experiment Video
Updated: Sep 20, 2025
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
Understanding Public Perceptions of Measles from Twitter Using Multi-Task Convolutional Neural Networks
Samuel Wang1, Jingcheng Du1, Lu Tang2
1School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, TX, USA.
Abstract:
Measles is a highly contagious cause of febrile illness typically seen in young children. Recent years have witnessed the resurgence of measles cases in the United States. Prompt understanding of public perceptions of measles will allow public health agencies to respond appropriately promptly. We proposed a multi-task Convolutional Neural Network (MT-CNN) model to classify measles-related tweets in terms of three characteristics: Type of Message (6 subclasses), Emotion Expressed (6 subclasses), and Attitude towards Vaccination (3 subclasses). A gold standard corpus that contains 2,997 tweets with annotation in these dimensions was manually curated. A variety of conventional machine learning and deep learning models were evaluated as baseline models. The MT-CNN model performed better than other baseline conventional machine learning and the signal-task CNN models, and was then applied to predict unlabeled measles-related Twitter discussions that were crawled from 2007 to 2019, and the trends of public perceptions were analyzed along three dimensions.
Insights
Public perception of measles, a contagious childhood illness, was analyzed using a novel AI model. The multi-task Convolutional Neural Network (MT-CNN) classified tweets to track public sentiment and vaccination attitudes over time.
Area of Science:
- Public Health
- Computational Linguistics
- Artificial Intelligence
Background:
- Measles is a highly contagious disease causing febrile illness, predominantly in young children.
- Recent years have seen a resurgence of measles cases in the United States, necessitating an understanding of public perception.
- Analyzing public discourse is crucial for effective public health responses to disease outbreaks.
Purpose of the Study:
- To develop and evaluate a multi-task Convolutional Neural Network (MT-CNN) model for classifying measles-related tweets.
- To analyze public perceptions of measles, including message type, expressed emotions, and attitudes towards vaccination.
- To track trends in public perception of measles and vaccination from 2007 to 2019.
Main Methods:
- A manually curated gold standard corpus of 2,997 measles-related tweets was annotated across three dimensions: message type, emotion, and vaccination attitude.
- A multi-task Convolutional Neural Network (MT-CNN) model was proposed and compared against conventional machine learning and single-task CNN models.
- The best-performing MT-CNN model was applied to a large dataset of unlabeled Twitter discussions from 2007-2019.
Main Results:
- The MT-CNN model demonstrated superior performance compared to baseline conventional machine learning and single-task CNN models.
- The model successfully classified measles-related tweets across message type, emotion, and vaccination attitude.
- Analysis of predicted unlabeled tweets revealed trends in public perception of measles and vaccination over a 13-year period.
Conclusions:
- The MT-CNN model is an effective tool for analyzing public perception of measles from social media data.
- Understanding public sentiment and vaccination attitudes is vital for informing public health strategies.
- The study provides insights into the evolving public discourse surrounding measles and vaccination.
