Computational Social Networks and Demand Forecasting

First, we aim to develop novel methods and algorithms to better understand, predict, support and control social behaviour in complex online systems such as e.g., social networks. Understanding the factors that govern, for example, consensus building processes in online collaboration systems, as well as underlying mechanisms that may turn such processes into a success or into a failure, enables us to implement intervention techniques (i.e., recommending experts as possible collaborators) with the goal to speed up the process of consensus building. With our approach we are able to address biases in online information systems such as differences in social status, polarization, discrimination and echo chambers.
Second, we aim to model user incentives by analyzing user historical data (i.e., in the automotive industry) and understanding rapidly changing market demands. By detecting certain trends, we are able to detect market saturations and forecast demands for the future.

Applications Publications People

Engineering Consensus Building in Complex Networks

Demand Forecasting