Konstantin is a PhD student in Urban Informatics and Data Science, based at the University of Warwick and affiliated with New York University. Konstantin is fascinated with the opportunities and challenges the AI revolution brings to the social sciences. With an undergraduate degree in Economics from University of Freiburg and a Masters in Transportation from Imperial College & UCL, he has gained interdisciplinary experience in tackling social questions using quantitative methods. He loves to actively engage with the community, for instance working for the Imperial College Data Science Society or organising a workshop at the Applied Machine Learning Days in Lausanne.
Konstantin’s research focuses on social problems with spatial dimensions, as they often occur in the urban domain. In particular, he is interested in leveraging the large data pools in cities to quantify and analyse urban structures and dynamics. This requires methods which:
- Offer the flexibility needed for high-dimensional, granular data.
- Provide a maximum of inference to not only predict social outcomes but understand the underlying processes.
- Are able to learn and account for spatial dependencies.
Models fulfilling all these needs are for instance Kernel Methods or Markov Random Fields. Two of Konstantin’s main applications are urban mobility and crime. Both problems are of a social nature and have an inherent spatial dimension. In his previous publications, he has used the urban built-environment to predict shared mobility activity and transfer this knowledge between cities.
Achievements and awards
- Winner of the CDRC Data Challenge 2018