Previous |  Up |  Next


Möbius transformation; expert systems; conditional weights; knowledge base
Möbius transformation is an important tool for establishing weights of compositional expert systems rules from conditional weights. In this paper, an applicability of Möbius transformation of rule bases is also extended to knowledge bases with elementary disjunctions in antecedents of rules. This paper contains an existence theorem, an algorithm of the transformation and some open problems which tend to maximal generality as well.
[1] Adlassnig K.-P.: Fuzzy set theory in medical diagnosis. IEEE Trans. Systems Man Cybernet 16 (1986), 2, 260–265 DOI 10.1109/TSMC.1986.4308946
[2] Daniel M.: A generalization of Möbius transformation. In: 4th Workshop on Uncertainty Processing (R. Jiroušek and G. D. Kleiter, eds.), Prague 1997, pp. 31–42
[3] Daniel M.: A Generalization of Möbius Transformation for Knowledge Bases which Include Rules with Disjunction in Antecedent. Technical Report No. V-706, ICS AS CR, Prague 1997
[4] Daniel M., Hájek P., Nguyen P. H.: CADIAG–2 and MYCIN–like systems. Artificial Intelligence in Medicine 9 (1997), 241–259 DOI 10.1016/S0933-3657(96)00376-4
[5] Hájek P.: Combining functions in consulting systems and dependence of premisses. In: Artificial Intelligence and Information – Control Systems of Robots (I. Plander, ed.), North–Holland, Amsterdam, 1984, pp. 163–166 Zbl 0559.68069
[6] Hájek P., Havránek T., Jiroušek R.: Uncertain Information Processing in Expert Systems. CRC Press, Boca Raton 1992 MR 1253068
[7] Hájek P., Valdés J. J.: An analysis of MYCIN–like expert systems. Mathware & Soft Computing 1 (1994), 45–68 MR 1301960
Partner of
EuDML logo