COVIDHealth - A Benchmark Twitter Dataset and Machine Learning based Web Application for Classifying COVID-19 Discussions

Aug 3, 2026

Upload by Violet

The COVID-19 pandemic has negatively affected both physical and mental health, prompting growing interest in extracting health-related insights from social media. This work develops a machine learning-based web application to automatically classify COVID-19 discussions, supported by labeled Twitter data, benchmark results, and a downloadable tool. Using the Twitter API, the authors label 6,667 tweets into five classes: health risks, prevention, symptoms, transmission, and treatment. Multiple feature extraction methods are combined with seven traditional algorithms and four deep learning models, reaching up to 90.43% F1 with CNN and 86.13% with Linear SVC.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages27

Reading Time1 h 8 min

AccessFree to download

Introduction

Motivation from Social Media for Public Health

Twitter as a Communication Channel

Need for COVID-19 Risk and Transmission Analysis

Study Contributions and Approach

arXiv :2402 .09897v1 [ cs .LG] 15 Feb 2024 COVIDHealth: A Benchmark Twitter Dataset and Machine Learning based Web Application for Classifying COVID-19 Discussions Mahathir Mohammad Bishala , Md. Rakibul Hassan Chowdorya , Anik Dasb , Muhammad Ashad Kabirc,∗ a Department of Computer Science and Engineering, Chittagong University of Engineering and Technology, Chattogram, 4349, Bangladesh b Department of Computer Science, St. Francis Xavier University, Antigonish, B2G 2W5, NS, Canada c Data Science Research Unit, School of Computing, Mathematics, and Engineering, Charles Sturt University, Bathurst 2795, NSW, Australia Abstract The COVID-19 pandemic has had adverse effects on both physical and mental health. During this pandemic, numerous studies have focused on gaining insights into health-related perspectives from social media. In this study, our primary objective is to develop a machine learning-based web application for automatically classifying COVID- 19-related discussions on social media. To achieve this, we label COVID-19-related Twitter data, provide benchmark classification results, and develop a web application. We collected data using the Twitter API and labeled a total of 6,667 tweets into five different classes: health risks, prevention, symptoms, transmission, and treatment. We extracted features using various feature extraction methods and applied them to seven different traditional machine learning algorithms, including Decision Tree, Random Forest, Stochastic Gradient Descent, Adaboost, K-Nearest Neighbour, Logistic Regression, and Linear SVC. Additionally, we used four deep learning algorithms: LSTM, CNN, RNN, and BERT, for classification. Overall, we achieved a maximum F1 score of 90.43% with the CNN algorithm in deep learning. The Linear SVC algorithm exhibited the highest F1 score at 86.13%, surpassing other traditional machine learning approaches. Our study not only contributes to the field of health-related data analysis but also provides a valuable resource in the form of a web-based tool for efficient data classification, which can aid in addressing public health challenges and increasing awareness during pandemics. We made the dataset and application publicly available, which can be downloaded from this link [https://github.com/Bishal16/COVID19-Health-Related](https://github.com/Bishal16/COVID19-Health-Related)Data-Classification-Website. Keywords: COVID-19 discussions, Twitter dataset, Deep learning, Machine learning, Classification, Web application 1. Introduction Social media platforms, including Twitter, Facebook, Whatsapp, Weibo, and others, have evolved into powerful channels for real-time communication during natural disasters and disease outbreaks across the globe [1] . These ∗Corresponding author [Email addresses:](Email addresses: mahathirbishal@gmail.com)[ mahathirbishal@gmail.com](Email addresses: mahathirbishal@gmail.com) (Mahathir Mohammad Bishal), [alvihasan361@gmail.com](alvihasan361@gmail.com) (Md. Rakibul Hassan Chowdory), [x2021gmg@stfx.ca](x2021gmg@stfx.ca) (Anik Das), [akabir@csu.edu.au](akabir@csu.edu.au) (Muhammad Ashad Kabir) Preprint submitted to arxiv February 16, 2024 platforms have become primary mediums for individuals to communicate, share their experiences, and exchange thoughts [2] . It holds the potential to serve as a valuable public health tool for scientists to promptly convey accurate information during pandemics, efficiently collecting reliable data [3] . Today, researchers harness the wealth of unstructured data from social media to construct effective frameworks for healthcare applications [4, 5] . Twitter, a microblogging and long-distance informal communication service, allows users to send “tweets" limited to 280 characters. With over 368 million monthly active users worldwide [6], it has become an essential platform for sharing ideas, data, and experimentation among medical experts for more than a decade [7, 8] . It has emerged as

FAQs

What is the main goal of the COVIDHealth study?

To build a machine learning-based web application that automatically classifies COVID-19-related Twitter discussions.

How many tweets and which classes are included in the labeled dataset?

A total of 6,667 tweets are labeled into five classes: health risks, prevention, symptoms, transmission, and treatment.

Which machine learning and deep learning models are used for classification?

Seven traditional algorithms are evaluated (e.g., Decision Tree, Random Forest, Logistic Regression, Linear SVC), and four deep learning models are tested (LSTM, CNN, RNN, BERT).

Aug 3, 2026

Upload by Violet