Related Experiment Video
Updated: Jul 29, 2025
Continuous Theta Burst Stimulation of the Posterior Medial Frontal Cortex to Experimentally Reduce Ideological Threat Responses
Published on: September 28, 2018
COVID-19 Vaccine Hesitancy: A Global Public Health and Risk Modelling Framework Using an Environmental Deep Neural
Miftahul Qorib1,2, Timothy Oladunni3, Max Denis4
1Department of Computer Science and Information Technology, University of the District of Columbia, Washington, DC 20008, USA.
Social media data, like tweets, can reveal public sentiment and emotions regarding vaccines. Analyzing these posts using advanced models like BERT helps identify factors influencing vaccine hesitancy for public health insights.
Area of Science:
- Computational Social Science
- Public Health Informatics
- Natural Language Processing
Background:
- Social media platforms are vital for rapid information dissemination and data collection.
- Analyzing public opinion on social media can inform public health strategies, particularly concerning vaccine hesitancy.
- Understanding public sentiment and emotions is crucial for addressing health-related concerns.
Purpose of the Study:
- To explore the utility of Twitter data for identifying factors contributing to vaccine hesitancy.
- To analyze public sentiment and emotions expressed in tweets related to vaccines.
- To evaluate the performance of various neural network architectures for classifying tweet sentiments and emotions.
Main Methods:
- Downloaded public tweets daily via the Twitter API.
- Preprocessed and labeled tweets using stemming, lemmatization, and the NRCLexicon technique for sentiment and emotion classification.
- Trained and tested 1DCNN, LSTM, Multiple-Layer Perceptron, and BERT models for multi-classification of sentiments and emotions.
Main Results:
- Statistical analysis revealed significant relationships between various emotion pairs (e.g., joy-sadness, trust-disgust) with p-values close to zero.
- Among the tested models, BERT achieved the highest accuracy of 96.71% in classifying COVID-19 related sentiments and emotions.
- 1DCNN and LSTM models also demonstrated high accuracy, achieving 88.6% and 89.93% respectively.
Conclusions:
- Social media analysis, particularly using advanced NLP models like BERT, is effective for understanding public sentiment and emotions related to vaccines.
- The findings highlight the potential of leveraging Twitter data to inform public health decision-making and combat vaccine hesitancy.
- BERT demonstrates superior performance in accurately classifying complex emotional and sentiment nuances in large-scale social media datasets.
More Related Videos
07:12Protocol for Data Collection and Analysis Applied to Automated Facial Expression Analysis Technology and Temporal Analysis for Sensory Evaluation
Published on: August 26, 2016
11:18Using the Threat Probability Task to Assess Anxiety and Fear During Uncertain and Certain Threat
Published on: September 12, 2014
Related Concept Videos
02:16Stereotype Content Model
01:51Vaccinations