Data mining and knowledge discovery in large databases are of significant importance in many Operations Research disciplines such as economics, finance, computational biology and life sciences. The progress in data acquisition and the recent technical advancements have resulted in the generation of massive databases in these fields, including for example credit card usage data, financial transactions data, supermarket transactions data, airline customer records and government statistics. The analysis of such large datasets requires sophisticated explorative methods for transforming experimental data into business intelligence. These approaches are currently used in a wide range of applications such as geo-marketing, market basket analysis, target marketing, customer relation management, but also in profiling practices like surveillance and fraud detection.
This course discusses the principle concepts from statistical data mining required for extracting patterns and conclusions from very large observational data sets. We present the main ideas in the field of clustering and classification and discuss its implications for data driven modeling and complex decision support. Finally, we demonstrate how data mining and game-theoretic approaches contribute to the identification and optimization of complex regulatory networks such as eco-finance networks and gene-environment networks.
Language: English
Duration: 10 academic hours
Tutor of the course:
Dr. E. Kropat, Institute for Theoretical Computer Science, Mathematics, and Operations Research, University of the Bundeswehr Munich, Germany