Did the coronavirus make Polish people angry? Or maybe it encouraged us to help others? How did isolation affect the crime rate? Data Science students from WUST investigated information from social media during the pandemic and reached interesting conclusions.

Data Science is an interdisciplinary field of science that consists in analysing data stored in computer systems. It uses mathematical, programming, computational, and analytical methods. Based on data analysis, it is possible to predict or model specific behaviours and events.

Students from the Faculty of Computer Science and Management decided to analyse users' activity in social media (e.g. Twitter and websites of public institutions) during the coronavirus-related lockdown.

– This was a joint initiative that came with the beginning of the pandemic – says Professor Przemysław Kazienko, head of the Data Science specialisation at the Faculty of Computer Science and Management. – We spontaneously added these activities to the curriculum as part of a student project – explains the professor.

He stresses that the analysis of social media to some extent reflects the current social moods. – Of course, with the reservation of representativeness, as not everyone is active in these media to the same extent – points out Professor Kazienko.

One group of students carried out a project called "Pandemotions" and the other - "The Crown Effect". They posted the results of their observations on publicly available research websites.

Optimism during the pandemic

Maciej Gruszczyński, Daniel Popek, Dawid Rusiecki, and Marcin Wątroba are the team that analysed over 300 thousand tweets in terms of mood and expressing emotions during the pandemic. They categorised their content as neutral, motivating or expressing optimism, helplessness, derision, fear, or anger.

– Every fifth tweet encouraged action and attention to common safety – say the students. Until the introduction of the first restrictions, negative emotions prevailed in the society. The turning point was the closure of educational institutions, at which point the tweets started becoming more and more optimistic. Instead of expressing anger, they motivated people to remain at home and stay calm. In May and June, anger started to prevail again.

It turned out that the subject of the virus was also discussed in a political context. The "Government", especially the "Ministry of Health", as well as the media and presidential candidates were quoted most often. On the day when the information about the postponement of the elections was announced, an exceptional outcry was observed among Internet users.

The popularity of #coronavirus, #covid19 and #stayathome is constantly decreasing, which is undoubtedly due to the lifting of restrictions and Poles’ return to normalcy.

– The machine learning algorithms we use help to detect emotions among Twitter users. We use five models, three of which are deep neural networks – explain the students.

The data presented is automatically updated using special tools. Their application downloads data related to the coronavirus from Twitter, and then an artificial intelligence module evaluates emotions which the tweets contain. After the prediction has been made, all graphs are updated. The results can be followed at http://pandemocje.pl.

The crown effect

The second team, comprising Joanna Komoszyńska, Zuzanna Świderska, Martyna Laszczyńska, and Michał Gołębiewicz, checked how the pandemic affects social life. The students analysed various sources of data, including the websites of the police, MPK Wrocław (municipal public transport company), the Gazeta Wrocławska newspaper, as well as the official Twitter profiles of the Ministry of Health and police.

They asked themselves questions including: Does the police have less work? How has the situation on the roads changed? Was Easter a quiet time? Is the work of journalists in the context of the pandemic the same as before? Is public transport working better? Did ministry announcements effectively reach Poles?

– The number of posts added by the Ministry of Health increased with the development of the situation in the country and the subsequent stages of restrictions. Over time, the average activity started to decrease, and after the holiday period, it remains relatively stable – observed the students.

They also included graphs showing the number of road accidents in 2020 compared to previous years, as well as the number of police interventions.

All analyses are available on the website.

Professor Przemysław Kazienko explains that there are many other areas in which IT can be used in the pandemic crisis. – For example, a large-scale study of the dynamics of change in relation to individual ideas or administrative solutions, such as a ban on leaving home –says Professor – Or control of the movement of individual persons based on location data, which is already working, or observations of social distancing based on video material recorded by cameras, e.g. in a store, or applications on phones that directly evaluate distance – explains the academic.

He stresses, however, that many IT solutions may raise doubts from the ethical point of view and should be considered in terms of social acceptance and long-term consequences.

More information about Data Science can be read on the website and the Facebook profile.