Arno Knobbe is a senior researcher in the LIACS Data Mining
group. The subject of his thesis, which can be found here, was Multi-Relational Data Mining. His current research includes the following subjects:
Multi-Relational Data Mining
MRDM is the art of
mining useful knowledge from structured data stored in relational databases.
Because structured data requires a database which contains multiple related
tables, traditional single-table techniques cannot be used. The field of MRDM is
concerned with generalising common Data Mining concepts to a multi-table
setting. Over five years of research in this area has resulted in the Safarii MRDM
system, which includes a whole range of Data Mining facilities organised around
a relational database containing multiple
tables.
Bioinformatics
Arno Knobbe is involved in the 'European Embryonal Tumor
Pipeline', an EU-funded research project concerning the genetic causes of
several cancers in small children. Along with the Jozef Stefan Institute and the
German Cancer Research Fund, he is responsible for the data analysis part of the
project, with a specific focus on the integration of knowledge from multiple
sources. His current activities involve the 'Systems Bioinformatics' project at LIACS.
Local Patterns and Pattern Teams
Most modern pattern
discovery algorithms produce an abundance of results. In many cases, the
end-user has limited time and is only interested in finding the few key patterns
in the data. The Pattern Team framework is concerned with finding these few non-redundant patterns within large collections
of local patterns. The different Pattern Team solutions have been implemented on
top of Safarii.
Learning in Computer Vision
A recent system developed by Arno Knobbe is the Computer
Vision system called Harmonii. The system uses
stereo-images to build up a 3D representation of the filmed scene, and plan
trajectories for mobile robots in an automotive setting. Machine Learning
algorithms are used to classify different areas of the image, based on colour,
texture, 3D orientation etc., in order to recognize features such as
road-surface en traffic signs.
A full list of publications since 1995 can
be found here.
