J2-2353 — Annual report 2009
1.
Semantic subgroup discovery: Using ontologies in microarray data analysis

We improved the SEGS algorithm (Searching for Enriched gene Sets) which finds subgroups of differentially expressed genes from experimental microarray data. For subroup discovery, SEGS uses background knowledge from domain specific ontologies GO, KEGG and ENTREZ.

COBISS.SI-ID: 22990631
2.
Advancing data mining workflow construction: A framework and cases using the Orange toolkit

We have developed an ontology of machine learning algorithms and annotated most of the algorithms incorporated in the Orange toolkit (developed at FRI, University of Ljubljana). We have also developed an algorithm for automatic datamining workflow construction, based on fast forward planner.

COBISS.SI-ID: 23502887