Twitter Interaction to Analyze Covid-19 Impact in Ghana, Africa from March to July

TL;DR

This study analyzes Twitter data from Ghana during the early COVID-19 pandemic to identify key topics, sentiment trends, and user engagement patterns from March to July 2020.

Contribution

It applies text mining to social media data to uncover public sentiment, topics, and engagement changes related to COVID-19 in Ghana, providing insights into societal reactions.

Findings

User engagement was highest in March and declined thereafter.

Key topics included COVID-19 information and social sentiments.

Sentiment analysis revealed changes in public mental state over time.

Abstract

The novel coronavirus, COVID-19, has impacted various aspects of the world from tourism, business, education, and many more. Like for every country, the global pandemic has imposed similar effects on Ghana. During this period, citizens of this country have used social networks as a platform to find and disseminate information about the infectious disease and also share their own opinions and sentiments. In this study, we use text mining to draw insights from data collected from the social network, Twitter. Our exploration of the data led us to understand the most frequent topics raised in the Greater Accra region of Ghana from March to July 2020. We observe that the engagement of users of this social network was initially high in March but declined from April to July. The reason was probably that the people were becoming more adapted to the situation after an initial shock when the disease was announced in the country. We also found certain words in these tweets of users that enabled us to understand the sentiments and mental state of individuals at the time.