Technique of Information Database Formation for Carrying Out Multilevel Monitoring and Classificatory-and-Forecasting Modelling

Abstract

The article offers a technique of computer information database formation for carrying out multilevel monitoring and classificatory-and-forecasting modelling. The author considers an algorithm of quality indices transformation in numerical values based on expert assessments, an algorithm of information filtering allowing to exclude doubtful messages, a modified algorithm ZET based on a missing values prediction accounting competent lines and columns of an initial database, an algorithm of the characteristics information value assessment, a procedure of the construction of integrated indices being an additive convolution of several unrelated local components according to their importance. The offered technique is recommended to be used for databases formation for carrying out medical-and-social monitoring and data intellectual analysis.