Data Mining and Knowledge Discovery (DM&KD) is a wide research area focusing upon methodologies for extracting useful knowledge from data. The ongoing fast growth of online data due to the Internet and the widespread use of databases have created an immense need for DM&KD methodologies.
Our R&D department and Sobolev Institute of Mathematics conduct fundamental scientific research in the field of Data Mining and Knowledge Discovery.
Our goal is to develop flexible program modules fit to any specific DM&KD task. It will allow to effectively check any hypotheses on various types of the data; to receive duly signals about the latent patterns and trends in the data; to represent results of data analysis in simple visual representations, etc. Thus, it will enable to have usable solution to Data Mining tasks, both for experts and for people without a deep knowledge of dynamic sequences of data.
Areas of expertise
- Knowledge Discovery and Analysis
- Pattern Recognition methods development
- Detection and verification of first-order logic hypotheses
- Data Mining on SQL databases
- Information integrity verification
- Case-based reasoning
- Chaotic and fractal decomposition
- Systematic methods of scientific inference
- Taxonomy and pattern recognition programs development
- Error searching and filling of gaps in charts
- Informative features selection in charts
- Programs for prognosis of multidimensional dynamic series