“Algorithms determine our world”: how the HSG trains students in complex data analysis

“Data are ubiquitous in today’s world. Data, some people say, are the new oil,” says Johannes Binswanger, Professor of Business Economics and Public Policy at the HSG. This wealth of data can be used by companies, NGOs and government organisations to make better decisions, says Binswanger. “However, many managers and decision-makers don’t know enough about how data analyses work, what resources they require and what kind of value can be derived from them.” This is one of the reasons why Binswanger set up the Data Science Fundamentals (DSF) certificate programme for HSG undergraduates in 2017.

The term “data science” denotes an interdisciplinary branch of science which filters out and visualises findings and patterns from data and thus paves the way for predictions. “The DSF programme is intended to help students bridge the gap between managers and data analysts,” says Binswanger. “In our world that is characterised by data, as many students as possible should have a basic knowledge of data science when they start their professional careers,” adds Lyudmila Grigoryeva, Associate Professor of Quantitative Methods for Economics. She teaches the initial unit of the DSF together with Binswanger.

Critical examination of algorithms

In this kick-off course of the DSF certificate programme – also called Data Science Bootcamp – students learn basic programming skills and how they can interpret existing data analyses. To begin with, there is an intensive two-week block seminar. Subsequently, students program an algorithm about an issue of their choice in groups, which is capable of making decisions on the basis of existing data sets. In the last few years, this has included algorithms for the prediction of forest fires, delays in public transport, the assessment of cancer risks, and private individuals’ creditworthiness.

The DSF courses other than the bootcamp can be freely compiled by students from a wide range of offers. The only compulsory subject is an introduction to the legal, political, ethical, managerial and financial aspects of algorithms. “Today, algorithms affect almost all areas of our lives. Therefore a wide variety of HSG disciplines present their view of algorithms,” says Binswanger. A critical examination of data science also requires that students learn to assess which algorithms are suitable for which problems – and what risks there are that predictions made by algorithms turn out to be wrong.

Data science for charitable applications

In the bootcamp, faculty members rely on a mixture of theory and practice: individual steps required to program an algorithm are discussed, after which students try these steps out in small case studies. “Basically, we merely present the theory that is necessary for the application,” says Binswanger. Various visiting speakers also familiarise students with practical fields of application in the bootcamp.

Thus representatives of the University of Warwick’s Data Science for Social Good programme present charitable applications: in 2022, students of this programme developed an algorithm which is capable of predicting child poverty for individual regions. This enables aid organisations to use their resources in a more effectively targeted way. “I also feel that many HSG students in the course want to use their knowledge for a good cause,” says Grigoryeva, who before she was appointed as a professor at the HSG was the director of the Data Science programme at the University of Warwick. She also considers it to be a function of the special DSF programme at the HSG to dispel students’ reservations about this field. “Students can attend the programme without any particular previous knowledge. Today, a great number of data science applications can be realised without any in-depth knowledge.”

Photo: Adobe Stock / Nicolas Herrbach