Final presentations in the internship seminar Data-driven Solutions for the Smart City Hamburg

22 February 2024

Photo: UHH

On the 14th of February, the final presentations in the internship seminar "Data-driven Solutions for the Smart City Hamburg", which is conducted in cooperation with the Hamburger Hochbahn AG, took place at the main building of the University of Hamburg. Three groups presented their results and developed prototypes.

All groups consisted of four to six students, and the first group examined how the automated evaluation of websites using the Open AI GPT4 models to extract specific information about organizations works by developing a web prototype and running numerous tests. The second group focused on the development of machine learning models for the prediction of delays in the subway system by analyzing and processing existing data, which they supplemented with weather data to train various models and to analyze the impact of different features. The use case for group three involved the analysis of the hvv switch points, which provide carsharing cars. For this, the students analyzed exemplary sensor data from the parking spots to specify the usage and capacity of all hvv switch points to evaluate and predict the potential of existing and new hvv switch points. In doing so, they also considered information about the urban surroundings regarding restaurants, social activities, shops, etc.

After each presentation, the findings, existing challenges, and future developments were discussed. Eva Bittner, Chikaodi Uba, and I were amazed to see how differently and efficiently the students worked on the use cases in interdisciplinary groups. Many thanks to Jan Krause, who established this cooperation on the side of HOCHBAHN and is developing it further with us, as well as to Christina Heidrich, David Claus, Normen Wiegand, Martin Gedlek, Olaf Petermann, and all the students, Ellis Steinbiß, David Faber, Malte Kießling, Kai Metzing, Lukas Krellenberg, Mica Langkat, Hans Hergen Lehmann, Lukas Jepsen, Jeremy Herbst, Henri-Leon Kordt, Liu Can, Elias Kakar and further.

We are looking forward to further collaboration and the next semester!