Sorry, you need to enable JavaScript to visit this website.

Since the onset of the Coronavirus (COVID-19) pandemic, “business as usual” has developed a drastically different meaning. From the stock market to day-to-day interactions, business students at Saunders College were given a new perspective on human behavior in light of large scale events. 

Quang Bui, Assistant Professor of Management Information Systems, is one of many professors that have incorporated the influence of COVID-19 into their curriculum. “In MGIS-355 Business Intelligence, students learn different techniques to analyze and make sense of big data,” said Bui. “This spring 2020 semester, as the coronavirus outbreak creates an unprecedented pandemic, students in MGIS-355 have been asked to conduct a sentiment analysis of 1,000 tweets from the World Health Organization (WHO) related to the outbreak. The object is to understand how the public responds to the WHO’s announcements surrounding the outbreak.”

Out of the 23 students in his class, Dr. Bui selected the work of Celeste Gambardella, a second-year software engineering student, as an exemplary study with “great details on the analysis and interesting findings.” 

Celeste’s executive summary states that, “Following the current news is extremely important now more than ever, as the world is looking for every way possible to overcome this global pandemic of COVID-19. The data being collected currently is more imperative to finding trends and further analysis.

“My task was to get a better grasp of how the public is responding to the outbreak of the coronavirus by analyzing tweets. To do so, I needed to use Twitter’s Application Program Interface (API), allowing me to read tweets to analyze them in Orange, an open-source data visualization, machine learning, and data mining toolkit. In particular, targeting tweets regarding the WHO that are related to the coronavirus outbreak.
coronavirus word cloud












Figure 14. Word cloud generated in Orange about coronavirus.

“By using Twitter’s API and Orange, I was able to understand how the public is responding to the outbreak of the coronavirus. Something that fascinated me was the word cloud generated above, Figure 14, which shows a lot of the concern that people are expressing right now. Subsequently, looking into different combinations of variables and subgroups, I was able to further my analysis regarding how people are reacting to COVID-19. In Figure 17, the Date and Emotion of the tweets are used to generate this Box Plot. When looking at the model, we can conclude that Surprise and Joy have the widest ranges, followed by Sadness and Fear. Meanwhile, the smallest ranges are Anger and Disgust.”

Figure 17. Box Plot of Date and Emotion for Preprocessed Text in Orange.

Celeste wrote, “Overall, COVID-19 has impacted people all around the world. People are being told to self-isolate themselves, leaving people to interact with others through social media now more than ever. Through the data analyzed through Twitter’s API, it is clear that the emotions of people are all over the place. People do not know what to expect every day as the virus continues to spread."

“The data I have collected could be used by leaders today speaking to the nation to help address these concerns people are having based on their emotions.”

Celeste’s work, along with the work of her classmates, is an essential part of experiential learning at Saunders College of Business. Working with cutting-edge technology on new and developing issues ensures that even in unexpected situations, students can grow their skills and put them into practice. 

Blog Tags: