In a 2012 article, the Harvard Business Review (HBR) predicted that “Data Scientist” would be “the sexiest job of the 21st century”. Since then, the prediction seems to have come true, as several Ivy League schools and other top ranking universities have scrambled to establish new institutes and Master’s programs in data science, while private bootcamps promise wealth and success to anyone who enrolls in their three month immersive programs in the nascent field.
Despite the buzz, data science is essentially an implementation of Statistics with the scale and automation that Computer Science can provide. This merging of disciplines is the product of the explosion in data collection of past two decades. It has bred a new field of practices and knowledge, obsessed with predicting the most minute details of our daily lives.
As predictive data analysis takes over an ever broader sphere of our lived experience, we find it urgent that more people improve their understanding of how these techniques are deployed and the logic that underlies them. We need this knowledge to resist and subvert the encroaching power of data. When and how can we deploy these techniques for our own purposes? Can we protect ourselves from predictive analytics by learning the very techniques it is based upon? How can we put data science to new and surprising uses? In organizing DeepMay, we aim to move beyond critique and hope to explore how data science can be used as a tool power bottom-up political initiatives.
Basics of data manipulation
During the course, we introduce students to the fundamentals of programming with the building blocks of any data science application: matrices, vectors and dataframes.
For there to be data science, there must also be data. We teach students how to obtain interesting datasets using web-scraping tools such as BeautifulSoup and Application Programming Interfaces (APIs) to access platforms and the data they offer.
Working with these kinds of real life datasets, we explore different visualization methods as well as the fundamentals of Geographical Information Science and mapping. How can you use visualizations to understand and present large and messy datasets? What are the different ways in which we can map our territories and different terrains of conflict, in order to increase our strategic capacities?
Intro to Machine Learning
Even though the primary emphasis on the bootcamp is on fundamentals as well as data collection and data exploration through visualization, we also visit the basics of machine learning. These basic techniques allow students to build simple predictive regression and classification models. Ultimately, we want students to really understand the tech stack that a data scientist is expected to know and in this way help them to continue
About the instructors