Treinei um algoritmo para seguir 80 bilhões de dinheiro público | Redes Neurais Ep. 1: Grafos

Did you know that it's possible to take public money and trace its path from your taxes to the company that ultimately received it, using only open data and a type of database that few people use daily? And I did it. I combined dozens of government databases into a map of over 1 million entities,

and in the end, 80 cases remained that, to say the least, demand explanation. Some are linked to people who were elected this year, now it's 2026. But hold on, before talking about who, because this isn't a political video, I need to show you how, because that's how you separate this from conspiracy theory, right? We use methodology. In

these three videos, the subject is technical. We're going to understand how a machine reads public money and how it can interpret the trail, the path, the route of this money. And we're going to use the case of public money in Brazil, because every day someone asks for more transparency in public funds. So I thought there was no

better way to talk about how to work with this technical solution than using public money itself. But one thing few people know is that the data has been public since 2004, when the government launched the transparency portal, and in 2009 it became law. It became law for municipalities, states, and the federal government. However, each

agency publishes its part in a different place, in a different format. And today we're going to learn one of the possible methods to combine this in an efficient and effective way to generate insights. And if this is the first time you've seen me here, I'm Maria Alice Maia, a data scientist from the University of California, Berkeley.