Sociology to prevent non-sense correlations on Covid

Inspiration

At the beginning, we wanted to see what kind of method was effective in reducing the pace of the infection, using simulation and big data. We then realized that the data that was provided was lacking some important aspect about the culture of the region studied, and that we could only get correlation rather than causation using these data

What it does

We then decided to use the best way to improve these data: a qualitative field study. Hence we want to promote the pre-project : this sociology part that engineers and informaticians always forget and devalue.

How I built it

What we will do concretely: We will interview people by skype or ZOOM now and in two months to get a better understanding of the evolution of the perception of the pandemic context and the system. We want to focus our study on how people respect the rules and the recommendations made by the confederation, and eventually find out important factors that were not used in the existing correlations, like adaptation time or lassitude. A final step would be to integrate the elements we discovered to an intelligent agent simulation, and to check if we get a better fit

Challenges I ran into

The time scale of the hackathon is not compatible with a comprehensive sociological study, and we would thus need to extend the project in time

Accomplishments that I'm proud of

We are quite proud to have been able to see beyond the classical view of hackathon, where the main aim is to build an app, and to dive into real solution finding issues.

What I learned

Hackathon format has a detrimental normative impact on its participants, that are pushed to lauche project without foundation or preliminaries checks. This format is efficient in computer science, but not to solve global issues like a pandemic, which require a more comprehensive view.