Modifications of algorithms of the automated diagnostic management system

The article is devoted to the task of modifying algorithms of the diagnostic management system for various spheres of the national economy. In previous studies, it was found that the task of managing diagnostics is important and necessary, while new effective methods are required to solve it, allowing objects to be assigned to a particular class with high accuracy, regardless of their internal component. At the same time, the basic requirements for diagnostic systems are put forward. The aim of the study. The purpose of this work is to construct modifications of the algorithms of the committee method for solving the problem of pattern recognition, as well as their illustration on a specific example of the diagnosis of sentinel headaches in stroke from others. Materials and methods. The modification of the algorithms of the committee method is based, on the representation of the system of inequalities in polar form, using the algorithm for constructing the minimum committee of the original system. The efficiency of finite-step algorithms based on reducing the dimension of the original problem, including the method of projecting onto a plane, is shown. The advantage of this method is its low computational complexity, as well as the content of the minimum number of members of the inequality system committee. Results. Modifications of the algorithms of the committees method are used for a specific task of diagnosing headaches, while the method proves to be effective, as evidenced by low recognition errors of the first and second kind in the test sample. It is also concluded, that the diagnostic task has a common structure in all areas, including technical, economic, psychological, forensic, and therefore the proposed algorithms can be applied to the specified number of tasks. There is also the possibility of using the committee method as a certain ensemble of methods, on the basis of which it is possible to create some automated diagnostic management system. Conclusion. The results obtained allow us to assert that the use of modified algorithms of the committee method allows for effective management of headache diagnostics.