Scientific Projects

STREAMLINE will address the competitive advantage needs of European online media businesses (EOMB) by delivering fast reactive analytics suitable in solving a wide array of problems, including addressing customer retention, personalized recommendation, and more broadly targeted services. STREAMLINE will develop cross-­sectorial analytics drawing on multi­‐source data originating from online media consumption, online games, telecommunications services, and multilingual web content. STREAMLINE partners face big and fast data challenges.

Our goal is to overcome the present state of the art (decision trees, logistic regression) in analyzing customer behavior and predicting risks for bank customers. We emphasize the deployment of open source analytics software. Our partner is OTP Bank.

Society has reached a point of no return, one that leaves us completely reliant on omnipresent ICT-mediated communication. Mobile and sensor-rich portable devices connect millions of humans with petabytes of data and numerous on-line services. However, tearing down the physical-digital barrier in a scalable fashion requires both radically novel algorithmic knowledge and in-depth understanding of humans and societies.

We develop data stream processing and predictive analytics tools for telecommunications data. One of our goals is a distributed Internet of Things data hub.

In a research project for Bosch, we develop scrap detection models based on the physical sensors of the manufacturing devices as well as logs of the production environment.