BIDSA, Bocconi Institute for Data Science, a New Research Center Is Born

WITH AN INTERDISCIPLINARY APPROACH, BIDSA PUTS TOGETHER MORE THAN 30 SCHOLARS FROM DIFFERENT DEPARTMENTS WHO, IN THEIR RESEARCH, ANALYSE LARGESCALE DATASETS AND MODEL THE UNDERLYING PHENOMENA. IGOR PRUENSTER IS THE DIRECTOR

Being a hub for Bocconi researchers that use data science in their research and a reference for the scientific community. This is the idea behind the Bocconi Institute for Data Science (BIDSA), an interdisciplinary research center inaugurated today with an international conference.
 
The conference speakers, Michael I. Jordan (University of California, Berkeley), Marc Mézard (École Normale Supérieure), Aldo Rustichini (University of Minnesota) and Riccardo Zecchina (Politecnico di Torino) are not only some of the leading scholars in the field, but also the perfect example of the variety of disciplines that the research center wants to work with. Jordan, one of the fathers of machine learning, works on the border between statistics and computer science. "In computer science, the growth of the number of data points is a source of complexity”, he says, “that must be tamed via algorithms or hardware, whereas in statistics, the growth of the number of data points is a source of simplicity in that inferences are generally stronger". Mézard and Zecchina, two statistical physicists, were recently awarded the prestigious Lars Onsager Prize with their colleague Giorgio Parisi for results that have applications in computational biology, in neural technologies, in optimization problems of complex networks and information theory. Rustichini applies the analysis of large amounts of data to neuroeconomics.
 
The center aims to promote and carry out theoretical and applied research involving modeling or analysis of large amounts of data. Interest in data science and demand for data scientists are on the rise all over the world and in every sector of the economy. BIDSA will support the training and education of students and junior researchers in fields where the use of advanced statistical and computational methodologies is essential.
 
The center counts over thirty fellows and affiliates, mostly from Bocconi and representing all the Departments of the University, with the exception of Legal Studies. "I’ve been working at the establishment of the center since I arrived at Bocconi, last February," says the director, Igor Pruenster. "The next step will be the involvement of PhD and MSc students, very important in ensuring the interdisciplinary nature of a research center. I’m designing a visiting student program, following the IGIER model".
 
The center is already managing a PRIN research project on modern bayesian nonparametric methods. "In the pipeline," continues the director, "we have a workshop on data science in marketing studies and a research program on the effectiveness of advertising billboards, measured by analyzing how pedestrians move".
 

by Fabio Todesco