Data visualization and information retrieval exploration based on COVID-19 data

Date

2021-05

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Second wave COVID-19 pandemics in the world coming soon. But the management of pandemic information that grows much faster than epidemic is a big challenge. Data visualization analysis during the outbreak provides vivid and clear information in making appropriate public health awareness. Nowadays, AI is becoming more and more popular and important in computer science field. I have a strong curiosity and desire to explore related technologies. In this work, I randomly selected COVID-19 related Weibo and WeChat Official Account, the two most popular Chinese social media platforms, posts from January 2020 to April 2020 are analyzed. Specifically, a kind of Information Retrieval model, TF-IDF (term frequency-inverse document frequency) is used to summarize the topics of posts. The classification of virus spread I made are contact transmission (within about 6 feet), droplet transmission, aerosol (tiny droplet) transmission based on CDC’s publication[1]. A total size of 5.72MB data was collected during the research time. I exclude Hubei Province from the national data because the central tendency is influenced too much by the extremum. So, Data from other 33 provincial-level administrative regions provide a more objective reflection of the China situation. The result of this study graphically displays the national average situation of the COVID-19 outbreak depended on quantitative and qualitative analysis of Chinese social media data. Public health awareness plays an important role for persuading people to take safety measures like wear mask, keep social distance, and home quarantine so that the epidemic was quickly brought under control. Future studies will continue to use social media data to predict other pandemic disease severity, evaluate effectiveness of correct guidance for people to respond to the epidemic.


Embargo status: Restricted to TTU community only. To view, login with your eRaider (top right). Others may request access exception by clicking on the PDF link to the left.

Description

Keywords

Data Visualization, Information Retrieval, COVID-19

Citation