Charakterystyka sieci Bayesa jako metody wnioskowania w warunkach niepełnych i niepewnych danych

• Autorzy: Suchacka G. , Zgrzywa A.

Warianty tytułu

EN

Characteristics of the Bayesian networks used for inferring in the conditions of incomplete and uncertain data

• Języki publikacji: PL

Abstrakty

PL

W artykule przedstawiono podstawowe informacje na temat sieci Bayesa oraz możliwości zastosowania tych sieci do wnioskowania w warunkach niepewnych i niepełnych danych. Na przykładzie prostej sieci przyczynowej omówiono strukturę sieci, sposób wnioskowania i obliczania prawdopodobieństw, a także podstawowe pojęcia. Zarysowano problemy związane z budowaniem modeli sieci, przedstawiono obszary zastosowań tej metody wnioskowania oraz narzędzia służące do tworzenia i modelowania sieci Bayesa. W celu zilustrowania omawianych zagadnień autorzy zaprojektowali przykładową sieć Bayesa i wykorzystali program MSBN do przeprowadzenia wnioskowania dla tej sieci.

EN

The article presents basic information on Bayesian Networks and possibilities of using this technology to inferring in the situations when data is incomplete and uncertain. Basic terms, network structure and the way of inferring and estimating possibilities are discussed on the basis of simple causal network. Furthermore, authors outline problems connected with building network models, they present fields of BN applications and the most popular tools for building and modeling Bayesian Networks. In order to illustrate the discussed issues, authors designed a hypothetical Bayesian Network and they used a program called MSBN to carry out the process of inferring for that network.

Słowa kluczowe

PL

sieć Bayesa; obliczanie prawdopodobieństw; budowanie modeli sieci

EN

Bayesian networks; estimating possibilities; building network models

• Wydawca: Oficyna Wydawnicza Politechniki Opolskiej
• Czasopismo: Zeszyty Naukowe. Informatyka / Politechnika Opolska
• Rocznik: 2001
• Tom: Vol. 282, z. 1
• Strony: 29--52
• Opis fizyczny: Bibliogr. 33 poz., schem., tab.

Twórcy

autor

Suchacka G.

• Katedra Automatyki, Elektroniki i Informatyki, Politechnika Opolska, Zakład Systemów Informacyjnych, Politechnika Wrocławska

autor

Zgrzywa A.

Bibliografia

• [1] CICHOSZ P.: Systemy uczące się, WNT, Warszawa, 2000.
• [2] CLARK L.A. et al.: Modeling Customer Survey Data, Case Studies in Bayesian Statistics, Vol. 4, Springer-Verlag.
• [3] GENOVESE C.R., SWEENEY J.A.: Functional Connectivity in the Cortical Circuits Subserving Eye Movements, Case Studies in Bayesian Statistics, Vol. 4, Springer-Verlag.
• [4] GIVENS G.H., ZEH J.E., RAFTERY A.E.: Implementing the Current Management Regime for Aboriginal Subsistence Whaling to Establish a Catch Limit for the Bering-Chukchi-Beaufort Seas Stock of Bowhead Whales, Report of the Int’l Whaling Commission, Vol. 46, 1996.
• [5] GOLDMAN A. et al.: Response of CD4 Lymphocytes and Clinical Con-sequences of Treatment Using ddI or ddC in Patients with Advanced HIV Infection, J. Acquired Immune Deficiency Syndromes and Human Retro-virology, Vol. 11, Nr 2, 1996.
• [6] O’HAGAN A., WELLS F.S.: Use of Prior Information to Estimate Costs in a Wewerage Operation, Case Studies in Bayesian Statistics, Springer- Verlag, 1993.
• [7] HEARST M.A.: Banter on Bayes: debating the usefulness of Bayesian approaches to solving practical problems, IEEE EXPERT, November/ December, 1997.
• [8] HEDBERG S.R.: Is AI going mainstream at last? A look inside Microsoft research, IEEE Intelligent Systems, March/April, 1998.
• [9] HORVITZ E., PAEK T.: A Computational Architecture for Conversation, Proceedings of the Seventh International Conference on User Modeling, Banff, Canada, June, 1999.
• [10] HORVITZ E.: Agents With Beliefs: Reflections on Bayesian Methods for User Modeling, Proceedings of the Sixth International Conference on User Modeling, Sardinia, Italy, June, 1997.
• [11] HORVITZ E.: Principles of Mixed-Initiative User Interfaces, Proceedings of CHI'99, ACM SIGCHI Conference on Human Factors in Computing Systems, Pittsburgh, PA, May, 1999.
• [12] HORVITZ E.: The Lumiere Project: Bayesian User Modeling for Inferring the Goals and Needs of Software Users, Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence, July, 1998.
• [13] HORVITZ E.: Uncertainty, Action and Interaction: In Pursuit of Mixed-Initiative Computing, Intelligent Systems, September/ October, 1999, IEEE Computer Society.
• [14] JENSEN F.V.: An Introduction to Bayesian networks, UCL Press Ltd, London, 1996.
• [15] KNIGHT J.R.: Modeling Real Estate Sales Patterns Using the Gibbs Sampler, Poster Presentation, Proceedings of the Fourth Annual Conference Case Studies in Bayesian Statistics, Springer-Verlag, New York, 1997.
• [16] PARMIGIANI G. et al.: Modeling Risk of Breast Cancer and Decisions about Genetic Testing, Case Studies in Bayesian Statistics, Vol. 4, B. P. Carlin et al., Springer-Verlag, New York.
• [17] PEARL J.: Probabilistic Inference in Intelligent Systems: Networks of Plausible Inference, Morgan Kaufmann, 1988.
• [18] PLAIN S.: Artificial Intelligence Gets Real - Bayesian Networks, Computer Shopper, August, 1997.
• [19] PRICE P.N., NERO A.V., GELMAN A.: Bayesian Prediction of Mean Indoor Radon Concentrations for Minnesota Counties, Health Physics, Vol. 71, Nr 6, 1996.
• [20] RASMUSSEN L.K.: Bayesian network for blood typing and parentage verification of cattle, Dina research report 38, Aalborg University, De-partment of Mathematics and Computer Science, 1995.
• [21] RASMUSSEN L.K.: BOBLO: an expert system based on Bayesian networks to blood group determination of cattle, Reset report 16, Research Centre Foulum, Denmark, 1995.
• [22] SAHAMI M., DUMAIS S., HECKERMAN D., HORVITZ E.: A Bayesian Approach to Filtering Junk E-mail, AAAI Workshop on Learning for Text Categorization, 1998.
• [23] THOMAS A.: Genotypie Inference with the Gibbs Sampler Population Management for Survival and Recovery: Analytical Methods and Strategies in Small-Population Conservation, Columbia University Press, New York, 1995.
• [24] ZGRZYWA A.: Analiza zastosowań sieci Bayesa do wnioskowania pro-babilistycznego, Raport serii SPR Nr 27, Zakład Systemów Informacyjnych, Politechnika Wrocławska, 2000.
• [25] Bayesian Belief Nets, http://www.cs.ualberta.ca/~greiner/bn.html.
• [26] Bayesian Network Resources, http://excalibur.brc.uconn.edu/~bavnet/
• [27] Bayes-On-Line, http://www.kbe.co.za/products/bol.htm
• [28] HUGIN, www.hugin.dk
• [29] Lumiere Bayesian Reasoning, User Modeling, and Automated Assistance, http://research.microsoft.com/~horvitz/lum.htm
• [30] Microsoft Belief Network (MSBN), http://www.research.microsoft.com/research/dtg/msbn/MSBNApp.htm.
• [31] Microsoft Product Support Services, http://support.microsoft.com/directorv/
• [32] Netica Application, http://www.norsvs.com/download.html http://www.norsvs.com
• [33] WinBUGS User Manual, http://www.mrc-bsu.cam.ac.uk/bugs/welcome.html