In the 34th episode of the "All Day Research" podcast, we welcome Prof. Dr. Heike Trautmann, who talks to our host Marvin Beckmann about data stream mining, machine learning and optimization.
Prof. Dr. Heike Trautmann studied statistics at the University of Dortmund and worked in business informatics at the University of Münster for ten years before joining us at Paderborn University in 2023. Here she is Professor of Machine Learning and Optimization in Computer Science. "The most comprehensive term in the field [...] is trustworthy artificial intelligence, which has a lot to do with reliability and robustness [...]," says Prof. Dr. Heike Trautmann, summarizing the largest area of her work.
In order to create such trustworthy artificial intelligence, a machine learning process is required, whereby researchers work a lot with textual and statistical data. They try to reduce this data stream to the essentials and then utilize the data obtained (data stream mining). "The big difference [to fixed data sets] is that it is a data stream, it is continuous. [...] It comes as it is measured [by sensors, for example]." Prof. Dr. Heike Trautmann explains the term. Such data can then come from social media platforms, for example. The challenge then lies in algorithmically recognizing campaigns and bots from the very large amount of changing data.
Listen to the 34th episode of the "All Day Research" podcast now, wherever podcasts are available or on the Institute's website.
Do you like the "All Day Research" podcast? Then please leave a rating.
Missed the last episode? Don't worry, all episodes can also be found on the Institute's website.
