Introduction to Computational Sciences - Einzelansicht

Description

The use of computers to study social phenomena has a long tradition in the social sciences. This course will offer students a first glance into how to make use of predictive modelling (i.e., specific machine learning models and the workflow on how to build such models) to study social phenomena. In addition to introducing the students to the field of predictive modeling, we will tackle the question of whether the predictability of life outcomes is feasible and/or desirable. Even in the era of big data, the predictability of individual and social outcomes (e.g., unemployment, mortality or health outcomes, criminality, etc.) remains challenging. We will employ simulation studies to empirically evaluate the limits of predictive algorithms and bring forward the network science perspective on complex systems, such as social networks. By the end of the course, the students should have a broad understanding of the technical aspects of building a predictive model, the ethical aspects of its deployment, and the potential limits in building such models when dealing with outcomes that result from complex social systems organized in dynamic networks.

Prerequisites

• Interest in machine learning
• Experience with R as a programming language (e.g., data management, descriptives, plotting, generalized linear model)