The sigma-if neural network as a method of dynamic selection of decision subspaces for medical reasoning systems

• Autorzy: Huk M.
• Języki publikacji: EN

Abstrakty

EN

To-date research in the area of applied medical artificial intelligence systems suggests that it is necessary to focus further on the characteristic requirements of this research field. One of those requirements is related to the need for effective analysis of multidimensional heterogeneous data sets, which poses particular difficulties when considering AI-suggested solutions. Recent works point to the possibility of extending the activation function of a perception to the time domain, thus significantly enhancing the capabilities of neural networks. This change results in the ability to dynamically tune the size of the decision space under consideration, which stems from continuous adaptation of the interneuron connection architecture to the data being classified. Such adaptation reflects the importance of individual decision attributes for the patterns being classified, as defined by the Sigma-if network during its training phase. These characteristics enable effective employment of such networks in solving classification problems, which emerge in medical sciences. The described approach is also a novel, interesting area of neural network research. This article discusses selected aspects of construction as well as training of Sigma-if networks, based on a sample problem of classifying Arabic numeral images.

Słowa kluczowe

EN

artificial intelligence; data mining; sigma-if neural network; decision space selection; non-destructive neural network pruning

PL

sztuczna inteligencja; wybieranie danych; sieci nuronowe sigma-if; wybór miejsca decyzji; nieniszczące neuronowe przycinanie sieci

• Wydawca: University of Silesia, Institute of Informatics, Computer Systems Department
• Czasopismo: Journal of Medical Informatics & Technologies
• Rocznik: 2004
• Tom: Vol. 7
• Strony: KB65--73
• Opis fizyczny: Bibliogr. 33 poz., rys.

Twórcy

autor

Huk M.

• Department of Computer Science, Wroclaw University of Technology

Bibliografia

• [1] DUCH W., JANKOWSKI N., Survey of neural transfer functions, Neural Computing Surveys, Vol. 2, pp. 163-212, Santa Cruz, 1999.
• [2] DUCH W., JANKOWSKI N., Transfer functions: hidden possibilities for better neural networks, Proc. 9th European Symposium on Artificial Neural Networks (ESANN), pp. 81-94, Université catholique de Louvain, Brugge 2001.
• [3] ANDREWS R., DIEDERICH J. TICKLE A.B. Survey and critique of techniques for extracting rules from trained artificial neural networks, Knowledge-Based Systems, Vol. 8, No. 6, pp. 373-389, Queensland, 1995.
• [4] CRAVEN M., SHALVIK J., Using neural networks for data mining, Future Generation Computer Systems, Vol. 13, pp. 211-229, Carnegie Mellon, 1997.
• [5] KASKI S., Data exploration using self-organizing maps, Acta Polytechnica Scandinavica, Mathematics, Computing and Management in Engineering Series, No. 82, pp. 57, Helsinki, 1997.
• [6] SETINO R., LOEW W.K., FERNN: An algorithm for fast extraction of rules from neural networks, Applied Intelligence, Vol. 12, pp. 15-25, Shrewsbury, 2000.
• [7] COHEN S., INTRATOR N., A hybrid projection based and radial basis function architecture: initial values and global optimization, Pattern Analysis & Applications, Vol. 5, No. 2, pp. 113-120, Springer-Verlag, London, 2002.
• [8] BLANZIERI E., Theoretical interpretations and applications of radial basis function networks, Technical Report DIT-03-023, Informatica e Telecomunicazioni, University of Trento, 2003.
• [9] MHASKAR H.N., MICCHELLI C. A., Approximation by superposition of sigmoidal and radial basis functions, Advances in Applied Mathematics, Vol. 13, No. 3, pp. 350-373, Orlando, 1992.
• [10] CHEN T., CHEN R., Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems, IEEE Transactions on Neural Networks, Vol. 6, No. 4, pp. 911-917, IEEE Press, 1995.
• [11] KAVZOGLU T., MATHER P.M., The use of feature selection techniques in the context of artificial neural networks, Proc. 26th Annual Conference of the Remote Sensing Society (CD-ROM), University of Leicester, Leicester, 2000.
• [12] FREITAS A.A. On rule interestingness measures, Knowledge-Based Systems, Vol. 12, No. 5, pp. 309-315, Elsevier Science, British Computer Society, 1999.
• [13] FREITAS A.A. Understanding the crucial role of attribute interaction in data mining, Artificial Intelligence Review, Vol. 16, No. 3, pp. 177-199, Kluwer Academic Publishers, Norwell, 2001.
• [14] HUK M., Określanie istotności atrybutów w zadaniach klasyfikacyjnych przez niedestruktywną eliminację połączeń w sieci neuronowej, Pozyskiwanie Wiedzy z Baz Danych, Vol. 975, pp. 138-147, Akademia Ekonomiczna, Wrocław, 2003.
• [15] PRECHELT L., Connection pruning with static and adaptive schedules, Neurocomputing, Vol. 16, No. 1, pp. 49-61, Elsevier Science, Karlsruhe, 1997.
• [16] PUI-FAI SUM J., Extended Kalman filter based pruning algorithms and several aspects of neural network learning, PhD Dissertation, Department of Computer Science and Engineering, Chinese University of Hong Kong, 1998.
• [17] ABELES M., Role of the cortical neuron: integrator or coincidence detector ?, Israel Journal of Medical Science, Vol. 18, No. 1, pp. 83-92, Weizmann Science Press, Jerusalem, 1982.
• [18] MAASS W., Bishop C., Pulsed neural networks, pp. 261-293, MIT Press, Cambridge, 1999.
• [19] MAASS W., Paradigms for computing with spiking neurons, In J.L. van Hemmen, J.D. Cowan, and E. Domany, editors, Models of Neural Networks, Early Vision and Attention, Vol. 4, chapter 9, pp. 373-402, Springer-Verlag, New York, 2002.
• [20] BOTHE S., LA POUTRE J., KOK J., Error-backpropagation in temporally encoded networks of spiking neurons, CWI Technical Report SEN-R0036, Stichting Mathematisch Centrum, Amsterdam, 2000.
• [21] MANGASARIAN O.L., SOLODOV M.V., Serial and parallel backpropagation convergence via nonmonotone perturbed minimization, Optimization Methods and Software, Vol. 4, pp. 103-116, Taylor & Francis Group, Stanford, 1994.
• [22] LUO Z.A., TSENG P., Analysis of an approximate gradient projection method with application to the backpropagation algorithm, Optimization Methods and Software, Vol. 4, pp. 85-102, Taylor & Francis Group, Stanford, 1994.
• [23] OHNO-MACHADO L., MUSEN M.A., Modular neural networks for medical prognosis: quantifying the benefits of combining neural networks for survival prediction, Connection Science, Vol. 9, No. 1, pp. 71-86, Taylor & Francis Group, Stanford, 1995.
• [24] LIU Y., YAO X., Neural networks for breast cancer diagnosis, Proc. 1999 Congress on Evolutionary Computation, pp. 1760-1767, IEEE Press, New York, 1999.
• [25] OSSEN A., ZAMZOW T., Segmentation of medical images using neural-network classifiers, Proc. First International Conference on Neural Networks and Expert Systems in Medicine and Healthcare, pp. 472-432, Plymouth, 1994.
• [26] ANDERSON C., CRAWFORD-HINES S., Learning expert delineations in biomedical image segmentation, Proc. Conference on Artificial Neural Networks In Engineering, pp. 657-662, St. Louis, 2000.
• [27] VEROPOULOS K., CAMPBELL C., SIMPSON J., The automated identification of tubercle bacilli using image processing and neural computing techniques, Proc. International Conference on Artificial Neural Networks, Vol. 2, pp. 797-802, Springer, Skövde, 1998.
• [28] VEROPOULOS K., Machine learning approaches to medical decision making, PhD Thesis, Department of Computer Science, University of Bristol, March 2001.
• [29] CIOS K.J., MOORE G.W., Uniqueness of medical data mining, Artifitial Intelligence in Medicine, Vol. 26, No. 1, pp. 1-24, Elsevier Science, 2002.
• [30] GAMBERGER D., LAVRAC N., KRSTATIC G., SMUC T., Inconsistency tests for patients records in a coronary heart disease database, Proc. First International Symposium on Medical Data Analysis, pp. 183-189, Springer, Frankfurt, 2000.
• [31] GAMBERGER D., LAVRAC N., GROSELI C., Experiments with noise filtering in a medical domain., Proc. 16th International Conference on Machine Learning, pp. 143-151, Jožef Stefan Institute, Bled, 1999.
• [32] KUKAR M., KONONENKO I., GROSELJ C., KRALJ K., FETTICH J., Analysing and improving the diagnosis of ischaemic heart disease with machine learning, Artificial Intelligence in Medicine, Vol 16, No. 1, pp. 25-50, Elsevier Science, 1999.
• [33] SHAHAR Y., CHENG C., Knowledge-based visualization of time-oriented clinical data. Proc. American Medical Informatics Association Fall Symposium, pp. 155-159, Orlando, 1998.