On Analytical Solutions to the Problems of Maintaining Local Consistency

One of the primary problems, arising in algebraic Bayesian networks, is the problem of checking and maintaining consistency of the knowledge pattern. It can be reduced to the linear programming problem, which methods of solving are well studied. However, acting as black box, this approach is ill-suited to solution of another important problem-research of the sensitivity of the probabilistic logical inference. In this work we prove the analytical representation of solutions of maintaining the local consistency problem for the knowledge pattern of small size and show the results of the experiment, comparing effectiveness of the solution using obtained formulae and simplex-method. The problem is being solved for the first time.