Knowledge representation and reasoning under uncertainty in expert systems - probabilistic graphical models

Head of the project: RNDr.Ing. Anna Slobodová, PhD.

The object of the project are theoretical, mathematical problems connected with the automation of the reasoning process under uncertainty in expert systems. From the view of applications, we are interested in uncertainty cases quantifiable with probability measure. We represent knowledge (including uncertainty) as a network of states and probabilities. The problem is to develop a model capable of representing complex systems changing over time and to model such systems under a unified framework to deal with both time and uncertainty.

• extension of the mathematical basis for the automation of the reasoning process under uncertainty in expert systems.The subject of the research is mainly probabilistic reasoning in dynamic probabilistic networks,

• representing uncertain knowledge and inference in dynamic probabilistic networks,

• producing a theoretical basis for both representation and inference in probabilistic temporal networks

Selected Publications:

• On a Special Case of Probabilistic Logic
  Slobodová A.
  Proc. 10th Israeli Symposium on Artificial Intelligence, Computer Vision and Neural Networks, Ramat Gan, Information Processing Association of Israel, 1993, pp. 81-86.

• On Some Problems of Reasoning with Probabilistic Knowledge
  Slobodová A.
  Proceedings of the 3rd Workshop on Uncertainty Processing in Expert Systems, University of Economics, Prague, 1994, pp. 243-252.

• Belief Updating
  Slobodová A.
  Proc. ISUMA- NAFIPS`95, 3rd Intern. Symp. Uncertainty Modeling and Analysis and Ann. Conf. North American Fuzzy Information Processing Society, IEEE Computer Society Press, Los Alamitos, California, 1995, pp. 715-719.

• Conditional Belief and Bayesian Conditioning
  Slobodová A.
  4th European Congress on Intelligent Techniques and Soft Computing. Proceedings, Aachen, Verlag Mainz, Wissenschaftsverlag, Vol.1, 1996, pp. 130-134.

• Conditional Multivalued Mapping and Belief
  Slobodová A.
  3rd European Congress on Systems Science Proc., Rome, Edizioni Kappa, 1996, pp. 501-505.

• Multivalued Extension of Conditional Belief Functions
  Slobodová A.
  Qualitative and Quantitative Practical Reasoning, 1st Int. Joint Conf. on Qualitative and Quantitative Practical Reasoning ECSQARU-FAPR'97 Bad Honnef, Germany, Springer Verlag, Lecture Notes in Artificial Intelligence, Vol. 1244 /D.M. Gabbay, R. Kruse, A. Nonnengart, H.J. Ohlbach /eds.// Proceedings, 1997, pp. 568 - 573.

• Random Sets and Conditional Belief Functions.
  Slobodová A.
  5th European Congress on Intelligent Techniques and Soft Computing. Proceedings , Aachen, Verlag Mainz, Wissenschaftsverlag, Vol.1, 1997, pp. 62-66.

• Generalized Graphical Models and Conditional Belief Functions.
  Slobodová A.
  6th European Congress on Intelligent Techniques and Soft Computing. Proceedings, Aachen, Verlag Mainz, Wissenschaftsverlag, Vol.1, 1998, pp. 158-162.

• Conditional Belief Functions and Valuation - Based Systems.
  Slobodová A.
  Second European Conference on Highly Structured Stochastic Systems, Pavia, Italy, 1999, pp. 271-273.

• Expected Utility Networks in Transferable Belief Model.
  Slobodová A.
  Computational Intelligence,Theory and Applications, International Conference, 7th Fuzzy Days Dortmund, Germany, Springer Verlag, Lecture Notes in Computer Science, Vol.2206 / Bernd Reusch /ed.// Proceedings, 2001, pp.800 - 806.

• A New Approach to Decision Making in Transferable Belief Model.
  Slobodová A.
  Neural Network World , Vol.11, No.6, 2001, pp. 651 - 659.