COVID-19 Public Sentiment Insights: A Text Mining Approach to the Gulf Countries

Public perspective of COVID-19 in five Gulf countries: findings from social media listening study

Abstract

Social media has been the primary source of information from mainstream news agencies due to the large number of users posting their feedback. The COVID-19 outbreak did not only bring a virus with it but it also brought fear and uncertainty along with inaccurate and misinformation spread on social media platforms. This phenomenon caused a state of panic among people. Different studies were conducted to stop the spread of fake news to help people cope with the situation. In this paper, a semantic analysis of three levels (negative, neutral, and positive) is used to gauge the feelings of Gulf countries towards the pandemic and the lockdown, on basis of a Twitter dataset of 2 months, using Natural Language Processing (NLP) techniques. It has been observed that there are no mixed emotions during the pandemic as it started with a neutral reaction, then positive sentiments, and lastly, peaks of negative reactions. The results show that the feelings of the Gulf countries towards the pandemic depict approximately a 50.5% neutral, a 31.2% positive, and an 18.3% negative sentiment overall. The study can be useful for government authorities to learn the discrepancies between different populations from diverse areas to overcome the COVID-19 spread accordingly.

Why This Study Matters

Study Design & Scope

• Geographical focus: Six Gulf countries—Saudi Arabia, UAE, Kuwait, Oman, Qatar, and Bahrain.

• Timeframe: Tweets from March 1 to April 28, 2020.

• Data collection: Used Tweepy and the Twitter API to gather ~60,000 tweets containing keywords like “corona,” “COVID-19,” etc. After cleaning (deduplication, spam filtering), the dataset was reduced to 33,496 tweets.

• Preprocessing steps: Removed non-Arabic words, hashtags, mentions, hyperlinks, emojis, punctuation, single characters; normalized Arabic orthography; eliminated stopwords; and applied stemming using ISRIStemmer.

• Sentiment analysis: Employed TextBlob_ar, an NLP tool tailored for Arabic, to tag each word as negative, neutral, or positive, computed polarity, and aggregated sentiments per tweet.

Key Insights

• Sentiment Distribution: Approximately 50.5% of tweets were neutral, 31.2% positive, and 18.3% negative.

• Country Variations: Oman exhibited the highest percentage of positive tweets (34.16%), while the UAE had the lowest (28.27%).

• Temporal Trends: The initial neutral sentiment shifted to positivity as likely reflecting optimism as cases and fatalities stabilized or declined. By mid-April, negative sentiment increased again, as rising case counts and fatalities likely fuelled concern and anxiety.

• Common Themes from word clouds include terms like “corona,” “virus,” “pandemic,” “new cases,” and country-specific keywords such as “curfew,” “health ministry” (Saudi Arabia), “Ramadan,” “application,” "Sheikh" (Qatar), and “ministry,” “crisis” (Kuwait).

Implications for Practice

• Policy and public health messaging: The predominance of neutral and positive sentiment offers an opportunity—governments should reinforce messaging, highlighting successes and easing anxieties, especially as initial fear subsided.

• Mental health initiatives: With nearly 1 in 5 tweets reflecting negative emotions, authorities should proactively offer mental health support, such as hotlines and counseling, to address evolving emotional stress.

• Combatting misinformation: Sustained informal and neutral discussion also opens paths for misinformation. Monitoring social media and deploying corrective campaigns can help curb false narratives.

Final Thoughts