Prototype-based rules and similarity-based methods

Objective: create logical rules based on similarity to prototypes, show that they are more general than fuzzy rules and provide an alternative to neurofuzzy system, develop similarity-based systems.

Participants: Wlodzislaw Duch, Marcin Blachnik, Karol Grudzinski, Krzysztof Grabczewski

Time: early work done in 2000 with Karol Grudzinski, more since 2003 with Marcin Blachnik, summarized in his PhD: Prototype-based rules and their connections with fuzzy systems with applications in data classification (in Polish), Technical University of Silesia 2007, followed by more advanced meta-learning based on composition of transformations.

Relevant Projects: Meta-learning and similarity-based methods.

Results: papers, talks, Matlab toolkit (Marcin): see papers with Marcin Blachnik on this list, and talks on this topic on this list

Main papers:

Duch W, Similarity based methods: a general framework for classification, approximation and association,
Control and Cybernetics 29 (4) (2000) 937-968

Duch W, Adamczak R, Diercksen G.H.F, Classification, Association and Pattern Completion using Neural Similarity Based Methods.
Applied Mathematics and Computer Science 10:4 (2000) 101-120

Duch W, Towards comprehensive foundations of computational intelligence. | PDF file.
In: W. Duch and J. Mandziuk, Challenges for Computational Intelligence. Springer Studies in Computational Intelligence, Vol. 63, 261-316, 2007.

Duch W, Setiono R, Zurada J.M, Computational intelligence methods for understanding of data | PDF file.
Proc. of the IEEE 92(5) (2004) 771- 805
Duch W, Grąbczewski K, Heterogeneous adaptive systems IEEE World Congress on Computational Intelligence, Honolulu, May 2002, pp. 524-529.

Duch W, Grudziński K, Prototype based rules - new way to understand the data. IEEE International Joint Conference on Neural Networks, Washington D.C. 14-18.07. 2001, pp. 1858-1863

More papers on this topic:

Blachnik M, Duch W, LVQ algorithm with instance weighting for generation of prototype-based rules.
Neural Networks 24(8), 824–830, 2011. DOI: 10.1016/j.neunet.2011.05.013.

Duch W, Maszczyk T, Grochowski M, Optimal Support Features for Meta-Learning.
Book chapter, in: Meta-learning in Computational Intelligence. Studies in Computational Intelligence. Eds: N. Jankowski, K. Grabczewski, W. Duch, Springer 2011, pp. 317-358.

Grochowski M, Duch W, Fast Projection Pursuit Based on Quality of Projected Clusters.
Lecture Notes in Computer Science Vol. 6594, pp. 89-97, 2011.

Maszczyk T, Grochowski M, Duch W. Discovering Data Structures using Meta-learning, Visualization and Constructive Neural Networks
In: Advances in Machine Learning II, Springer Studies in Computational Intelligence, Vol. 262, pp. 467-484, 2010.
Duch W, Neurocognitive Informatics Manifesto.
In: Series of Information and Management Sciences, California Polytechnic State University, 8th Int Conf on Information and Management Sciences (IMS 2009), Kunming-Banna, Yunan, China, pp. 264-282.
Not directly related but some remarks on neurobiological connections of prototype-based rules are made.
Blachnik M, Duch W, Building Localized Basis Function Networks Using Context Dependent Clustering
Lecture Notes in Computer Science, vol. 5163, 482-491, 2008.
Presented at: International Conference on Artificial Neural Networks (ICANN'08), Prague, Czech Republic

Blachnik M, Duch W, Prototype rules from SVM | Abstract and PDF file.
Book chapter, in: Rule Extraction from Support Vector Machines, ed. J. Diederich, Springer Studies in Computational Intelligence, Vol. 80, 163-184, 2008.

Wieczorek T, Blachnik M, Duch W. (2008), Heterogeneous distance functions for prototype rules: influence of parameters on probability estimation. | Abstract.
International Journal of Artificial Intelligence Studies (in print, since 2006).
Blachnik M, Duch W Prototype-based threshold rules. | PDF file.
Lecture Notes in Computer Science, Vol. 4234, 1028-1037, 2006.
Blachnik M, Duch W, Wieczorek T, Selection of prototypes rules – context searching via clustering. | PDF file.
Lecture Notes in Artificial Intelligence, Vol. 4029, 573-582, 2006
Duch W, Rules, Similarity, and Threshold Logic.
Commentary on Emmanuel M. Pothos, The Rules versus Similarity distinction.
Behavioral and Brain Sciences Vol. 28 (1): 23-23, 2005

Blachnik M, Duch W, Wieczorek T, Probabilistic distance measures for prototype-based rules. | PDF file.
Proc. of the 12th Int. Conference on Neural Information Processing (ICONIP'2005), Taipei, Taiwan, pp. 445-450

Blachnik M, Duch W, Wieczorek T, Threshold rules decision list. | PDF file.
In: T. Burczyński et al. (Eds), Methods of Artificial Intelligence, AI-METH Series, Gliwice 2005, pp. 23-24

Wieczorek T, Blachnik M, Duch W. (2005), Influence of probability estimation parameters on stability of accuracy in prototype rules using heterogeneous distance functions. | PDF file.
In: Proceedings of the Artificial Intelligence Studies, Vol.2 (25), 2005, pp. 71-78.
Duch W, Blachnik M, Fuzzy rule-based systems derived from similarity to prototypes.
Lecture Notes in Computer Science, Vol. 3316 (2004) 912-917.
Grąbczewski K, Duch W, Heterogenous forests of decision trees. Springer Lecture Notes in Computer Science Vol. 2415 (2002) 504-509.
Duch W (1996) Categorization, Prototype Theory and Neural Dynamics, 113 KB.
Proc. of the 4th International Conference on SoftComputing'96, Iizuka, Japan, ed. T. Yamakawa and G. Matsumoto, pp. 482-485 (invited paper, cognitive science session)

Open problems: applications, theory.

Back to Dept. of Informatics list of projects