Inductive Rule Learning
Joint bilateral Slovenian-Croatian research project
2009-2010
The topic of joint research is the development and use of machine learning methods, enabling the discovery of new knowledge and relationships from data in the form of rules. The project is continuation of the previous bilateral projects "Intelligent Data Analysis" and
"Intelligent Subgroup Discovery".
The goals are development of rule induction systems, generalisation of the results in the area of inductive subgroup discovery, and formalisation of the area named supervised descriptive rule induction. The demonstration of the applicability of the developed methods will be through applications in medicine and systems biology. The project should help to establish long term connections among researchers, especially young people, enable preparation of joint
international projects and stimulate industrial applications of intelligent data analysis.
Project partners
Researchers
- dr. Nada Lavrac, Jozef Stefan Institute, project leader
- dr. Igor Mozetic, Jozef Stefan Institute
- dipl. ing. Petra Kralj, Jozef Stefan Institute
- dipl. ing. Vid Podpecan, Jozef Stefan Institute
- dr. Dragan Gamberger, Rudjer Boskovic Institute, project leader
- dr. Tomislav Smuc, Rudjer Boskovic Institute
- dipl. ing. Nives Skunca, Rudjer Boskovic Institute
- dipl. ing. Fran Supek, Rudjer Boskovic Institute
Present activities
- Public web service: Saturation filter for detection and elimination of noise in datasets.
- Contrast set mining and descriptive induction.
- Theory and practice of explicit noise detection in rule learning.
- Preparion of the book titled: "Rule Learning: Essentials of Machine Learning and Relational Data Mining".
Published papers
- Kralj, P., Lavrac, N., Gamberger, D. and Krstacic, A. (2009)
CSM-SD: Methodology for contrast set mining through subgroup discovery.
Journal of Biomedical Informatics 42/1:113-122.
Organized meetings
Last updated March 2009.