In order that the probabilistic logic reasoning under uncertainty can be used for 1arge scale knowledge-base,this paper presents a clustering-algorithm for decomposi-tion of probabilistic logic formula set based on the general consistence assigning algorithm for Drobabilistic logic and the experience of developing the knowledge-base in an practical expert system. It ensures that the united-model of Dantzig-Wolfe decomposition can be used for probabilistic logic reasoning on probabilistic logic knowledge-base with different background,provided the formula set hold certain hierarchical structure.Experiments show that the algorithm performs successfully on instances with dozens of variables and clauses.
1 Nilsson Nils J.Probabilistic logic.Artificial Intelligence,1986,28(1):71~87.
2 Genesareth M.Nilsson N.Logic fouladations for AI.Morgan Kaufmann,1 987.
3 Andersen K A.Characterizing censistency in probabilistic logic for a class of Horn clauses.Mathematical Program-
ming,1994,66:257~271.
4 Anderset K A,Hooker J N.Bayesian logic.Decision Support Systems,1994,11:191~210.
5 Kavvadias Dimitris.Papadimitriou Christos H.A linear programming approach to reasoning about probabilities.
Annals of Mathematics and AI,1990,1:189~205.
6 BOUChaffra D.A relation between isometrics and the relative consistency concept in probabilistic logic.In:Proceed—
ings of the 13th IMACS World Congress on Computation and Applied Mathematics, Dublin,July 1991.22~26.
7 Bouchaffra D.Consistent regions in probabilistic logic when using different norms.Proceedings of the Third Work-
shop on Artificial Intelligence and Statistics,Miami,Fort Lauderdale,FL,January 1991.5.1~5.5.
8 高文豪.大系统最优化.水利电力出版社,1991.
9 Dantzig G B.Linear programming and extensions.Princeton University Press,1963.
10 Dantzig G t3.Wolfe P.Decomposition principle for linear programming.Oper.Res.,1960·8:101~111.
11 张勇传,瞿继恂.组合最优化——计算机算法和复杂性.武汉;华中理工大学出版社,1994.
12 陈尚勤,魏鸿骏.模式识别理论及应用.成都:成都电讯工程学院出版社,1985.