Incomplete measurement data in the problem of classification

• Autorzy: Adamus E.
• Języki publikacji: EN

Abstrakty

EN

In the article the author has attempted to supplement, the incomplete measurement samples with an introductory analysis of the profitability of the possible prediction of the missing values. In order to do that a fuzzy model, automatically tuned by a self-organising neural network, has been applied. Thus, the method may be used for solving problems connected with classification. However, the idea of treating the incomplete data has been adopted from the rough set theory.

Słowa kluczowe

EN

incomplete input vector; neural gas algorithm; inconsistent rules

• Wydawca: Instytut Telekomunikacji i Informatyki Uniwersytetu Technologiczno-Przyrodniczego w Bydgoszczy
• Czasopismo: Image Processing & Communications
• Rocznik: 2005
• Tom: Vol. 10, no 2
• Strony: 85--92
• Opis fizyczny: Bibliogr. 20 poz., wykr.

Twórcy

autor

Adamus E.

• Department of Artificial Intelligence and Mathematical Methods, Faculty of Computer Science and Information Systems, Tedmical University of Szczecin, Żołnierska 49, PL-71210 Szczecin,,

Bibliografia

• [1] E. Adarnus 2003. "Przegląd metod stosowanych do badań nad niekompletnymi danymi pomiarowymi". VIII Sesja Informatyki. 2003, Szczecin.
• [2] G. Batista, and M. C. Monard. An analysis of four missing data treatment methods for supervised learning", Applied Artificial Intelligence, p. 519:533, 2003.
• [3] R. Berthold and K. P. Huber. "Missing values and learning of fuzzy rules", 1998
• [4] A. P. Dempster, N. M. Laird, and D. B. Rubin. "Maximum likelihood from incomplete data via the em algorithm". Journal of Royal Statistical Society, B39:l 38, 1977.
• [5] Z. Ghahramani and M. I. Jordan. Learning from incomplete data", AIM-1509, p. 11, 1994.
• [6] Z. Ghahramani and M. I. Jordan. Supervised learning from incomplete data via an EM approach". In Jack D. Cowan, Gerald Tesauro, and Joshua. Alspector, editors, Advances in Neural Information Processing Systems, volume 6, pages 120127. Morgan Kaufmann Publishers, Inc., 1994.
• [7] J. W. Grzymala-Busse. On the unknown attribute values in learning from examples". Proc. of the ISMIS-91, 6th International Symposium on Methodologies for Intelligent. Systems, Charlotte, North Carolina, Lecture Notes in Articial Intelligence, 542:368 377, October 16-19 1991.
• [8] J. W. Grzymala-Busse. "Knowledge acquisition under uncertainty - a rough set approach."Journal of Intelligent & Robotic Systems 1, 1988, 3-16.
• [9] J. W. Grzymala-Busse. Zough Set Strategies to Data with Missing Attribute" Values. Proceedings of the Workshop on Foundations and new Directions in Data Mining, associated with the third IEEE Interantional Conference on Data Mining, November 19-22, 2003, Melbourne, FL, USA, 56-63.
• [10] G. King, J. Honaker, and A. Joseph. Analyzing incomplete political science data": An alternative algorithm for multiple imputation. 2000.
• [11] W. Z. Liu, A. P. White, S. G. Thompson, and M. A. Brarner. "Techniques for dealing with missing values in classication". Lecture Notes in Computer Science, 1280, 1997.
• [12] Z. Pawlak. Rough sets, Int. J. Computer and Information Sci. 11, 1982, 341-356.
• [13] J. R. Quinlan. "Unknown attribute values in induction". In Proceedings of the sixth International Conference on Machine Lerarriing, 1989.
• [14] J. Schafer and M. Olsen. "Multiple imputation for multivariate missing-data problems: a data analyst's perspective". Multivariate Behavioural Research., Vol. 33, pp. 545-571, 2001.
• [15] A. Szymkowiak, P. A. Philipsen, and J. Larsen. "Imputating missing values in diary records of sun-exposure study", Conference Proceedings of (IEEE) Workshop on Neural Networks for Signal Processing (XI). pp. 489-498, 2001.
• [16] H. Timm. "Methoden zur exploration von daten niit fehlenden werten sowie classiflzerten daten". Doctoral Thesis, Faculty of Computer Science arid Information Systems, Technical University of Magdeburg, 2002.
• [17] V. Tresp and S. Ahmad. Some solutions to the missing feature problem in vision". Advances in Neural Information Processing Systems 5., 1993.
• [18] V. Tresp, S. Ahmad, and R. Neuneier. "Trainig neural networks with deficientdata". Advances in Neural Information Processing Systems 6., 1994.
• [19] V. Tresp, S. Ahmad, and R. Neuneier. Efficient methods for dealing with missing data in supervised learning". Advances in Neural Information Processing Systems 7., 1995.
• [20] S.-Y. Yoon and S.-Y. Lee. "Training algotithm with incomplete data for feed forward neural networks, volume Neurals Processing Letters 10: 171-179. Kluwer Academic Publishers, 1999.