Methods for Developing Hierarchical Models

Project Results

Publications

Theory and methodology

• Zupan, B., Bohanec, M., Bratko, I., Demšar, J.: Machine learning by function decomposition, in Proceedings ICML-97 (ed. Fisher, D.H.), pp. 421-429, San Francisco: Morgan Kaufman Publishers, 1997.
• Zupan, B., Bohanec, M., Demšar, J., Bratko, I.: Learning by discovering concept hierarchies, Artificial Intelligence 109, pp. 211-242, 1999.
• Zupan, B., Bratko, I., Bohanec, M., Dem{ar, J.: Induction of concept hierarchies from noisy data, in Proceedings of the Seventh International Conference on Machine Learning ICML-2000 (ed. Langley, P.), pp. 1199-1206, San Francisco: Morgan Kaufmann Publishers, 2000.

Partition selection measures

• Zupan, B., Bohanec, M.: Experimental evaluation of three partition selection measures for decision table decomposition, Informatica 22, pp. 207-217, 1998.

Handling of noise and uncertainty

• Zupan, B,: Machine learning based on function decomposition, Ph.D. Thesis, University of Ljubljana, 1997.

Decomposition of real-valued functions

• Demšar, J., Zupan, B., Bohanec, M., Bratko, I.: Constructing intermediate concepts by decomposition of real functions, in Machine Learning: ECML-97 (eds. van Someren, M., Widmer, G), pp. 93-107, Berlin: Springer-Verlag, 1997.

Applicability of function decomposition

For Decision Support:

Bohanec, M., Zupan, B., Bratko, I., Cestnik, B.: A function-decomposition method for development of hierarchical multi-attribute decision models, in Proceedings of the Fourth Conference of the International Society for Decision Support Systems ISDSS-97, pp. 503-514, Lausanne, 1997.

For Knowledge Discovery in Databases (KDD):

Zupan, B., Bohanec, M., Bratko, I., Cestnik B.: A dataset decomposition approach to data mining and machine discovery, in Proc. of the Third International Conference on Knowledge Discovery and Data Mining (KDD-97) (eds. Heckerman, D., Mannila, H., Pregibon, D., Uthurusamy, R.), pp. 299-303. AAAI Press, 1997.

For Feature Transformation:

Zupan, B., Bohanec, M., Demšar, J., Bratko I.: Feature transformation by function decomposition, IEEE Intelligent Systems 13(2), pp. 38-43, 1998. (abstract).

Zupan, B., Bohanec, M., Demšar, J., Bratko I.: Feature transformation by function decomposition, in Feature extraction, construction and selection: A data mining perspective (eds. Liu, H., Motoda, H.), pp. 325-340, Kluwer Academic Publishers, 1998.

Applications in Medicine:

Zupan, B., Halter, J.A., Bohanec, M.: Concept Discovery by Decision Table Decomposition and its Application in Neurophysiology, in Intelligent data analysis in medicine and pharmacology (eds. Lavrač, N., Keravnou, E., Zupan, B.), pp. 261-277, Kluwer, 1997.

Bohanec, M., Zupan, B., Rajkovič, V: Hierarhični odločitveni modeli in njihova uporaba v zdravstvu, Zbornik CADAM-97: Računalniška analiza medicinskih podatkov (eds. I. Kononenko, T. Urbančič), Institut Jožef Stefan, 1-17, 1997.

Applications in Socioeconomic Research:

Krisper, M., Zupan, B.: Synthesis of hierarchical decision support models from socioeconomic data, Zbornik konference Informacijska družba (eds. Bavec, C., Gams. M.), Institut Jožef Stefan, 60-63, 1998.

Methods and tools for hierarchical decision models

• Bohanec, M., Cestnik, B., Rajkovič, V.: Evaluation models for housing loan allocation in the context of floats, in Context sensitive decision support systems (eds. Berkeley, D., Widmeyer, G.R., Brezillon, P., Rajkovič, V.), Chapman & Hall, pp. 174-189, 1998.
• Grohar, B., Bohanec, M., Rajkovič, V.: DexW: Računalniški program za delo s kvalitativnimi večparametrskimi odločitvenimi modeli, Zbornik sedme Elektrotehniške in računalni{ke konference ERK'98 (ed. Zajc, B.), Portorož, 107-110, 1998.

Developed Software

After the termination of this project, HINT has been incorporated into Orange, a public domain data mining software developed at the University of Ljubljana, Faculty of Computer and Information Science.