Knowledge Management & Discovery Lab
KMD stands for "Knowledge Management and Discovery" .
The KMD Lab is part of the department Technical and Business Information Systems (ITI) .
The KMD Lab has been established in February 2003.
In the KMD lab, we develop and apply data mining methods for dynamic environments:
- to understand the progress of diseases and the long-term impact of interventions
- to make recommenders adaptive to changing user interests and market conditions
- to monitor opinions
- to learn actively from the data, minimizing human effort.
Our methods are mostly in the field of stream mining. We develop stream mining methods, methods that exploit timestamped data, and dedicated stream algorithms for recommendation engines, opinion analysis, patient records and longitudinal epidemiological data.
Our research is reflected in our teaching programs. With KMD, students learn fundamentals of data mining and recommendation engines. They learn to design and apply mining and machine learning methods in realistic applications, and get involved in our research - in team projects and individual projects.
All our projects are listed here.
Anne Rother's first-author paper, "Assessing the difficulty of annotating medical data in crowdworking with help of experiments", has been accepted in PLOS ONE. It is Anne's first scientific paper, which she wrote while still studying for her bachelor's degree. Congratulations on this great success!
Anne Rother, Uli Niemann, Tommy Hielscher, Henry Völzke, Till Ittermann, and Myra Spiliopoulou (2021). Assessing the difficulty of annotating medical data in crowdworking with help of experiments. PLOS ONE 16(7): e0254764. https://doi.org/10.1371/journal.pone.0254764
We are happy that Clara Puga's paper "Discovery of Patient Phenotypes through Multi-layer Network Analysis on the Example of Tinnitus", has been accepted at the IEEE International Conference on Data Science and Advanced Analytics 2021 (DSAA), which takes place in Porto, Portugal, 06-09 October. It is Clara's first scientific paper - Congratulations to her on this big achievement!
Clara Puga, Uli Niemann, Vishnu Unnikrishnan, Miro Schleicher, Winfried Schlee, and Myra Spiliopoulou: "Discovery of Patient Phenotypes through Multi-layer Network Analysis on the Example of Tinnitus"
Uli Niemann successfully defended his Phd thesis titled „Intelligent Assistance for Expert-Driven Subpopulation Discovery in High-Dimensional Timestamped Medical Data“.
Prof. Myra Spiliopoulou and Uli Niemann are organizing a workshop on "Mining and Policy-Making for Epidemic Surveillance" (EpiMine), together with Prof. Panagiotis Papapetrou and Maria Bampa from the University of Stockholm, Sweden.
The goal of EpiMine is to promote research in the areas of knowledge discovery, data mining, and policy-making that contribute to the realization and further development of effective and timely prevention measures and strategies to contain epidemics such as COVID-19.
The workshop will be held on 07 December 2021 in conjunction with the IEEE International Conference on Data Mining (ICDM) in Auckland, New Zealand.
More information on the workshop's webpage.
We are pleased to announce that our four papers have been accepted for the IEEE CBMS International Symposium on Computer-Based Medical Systems, 07-09 June 2021. Congratulations to all authors!
Circadian Conditional Granger Causalities on Ecological Momentary Assessment Data from an mHealth App by Noor Jamaludeen, Vishnu Unnikrishnan, Ruediger Pryss, Johannes Schobel, Winfried Schlee and Myra Spiliopoulou
Juxtaposing 5G Coronavirus Tweets With General Coronavirus Tweets During the Early Months of Coronavirus Outbreak by Rafi Trad and Myra Spiliopoulou
Love thy Neighbours: A Framework for Error-Driven Discovery of Useful Neighbourhoods for One-Step Forecasts on EMA data by Vishnu Unnikrishnan, Yash Shah, Miro Schleicher, Carlos Fernandez-Viadero, Mirela Strandzheva, Doroteya Velikova, Plamen Dimitrov, Rüdiger Pryss, Johannes Schobel, Winfried Schlee and Myra Spiliopoulou
User-centric vs whole-stream learning for EMA prediction by Saijal Shahania, Vishnu Unnikrishnan, Rüdiger Pryss, Robin Kraft, Johannes Schobel, Ronny Hannemann, Winfried Schlee and Myra Spiliopoulou
Noor Jamaludeen, Clara Puga and Anne Rother will participate in the virtual conference Women in Data Science (WiDS) Regensburg (13.+14.04.2021).
We will present the following topics:
"A Comparison of Model-Based Methods for Imputing Incomplete Multivariate Time Series" (Noor Jamaludeen)
"Data Science applied to Medical Research" (Clara Puga)
"Triplet-based-learning with the help of crowdlabeling on medical data" (Anne Rother)
Prof. Myra Spiliopoulou participates in the "Roadshow on Funding possibilities in the field of digitalization and industrial technologies (15 April 2021)" and reports on the Horizon 2020 Project UNITI under the title "UNITI - Medical research and Data Science jointly against tinnitus" (event in German)