The problem of the large-scale management system study is considered. The system consists of a large number of objects, each of which is characterized by a heterogeneous set of parameters. In the work, to solve the set tasks, it is proposed to investigate the structure of the relative location of these objects in the informative parameters space. This allows to significantly increase the analysis efficiency of the system functioning and the stability of the procedures for making management decisions. To identify such patterns special mining complicated data algorithms complex and expert correction procedures were designed. The complex includes the following algorithms: structural-classification data analysis (SCDA), selection of informative parameters, select the initial conditions for classification algorithms, select the "optimal" number of classes, completing missing observations, as well as procedures of expert correction of the results for these algorithms. The theoretical analysis of different variants SCDA algorithm was carry out, the algorithm convergence theorems to the local extremum of the appropriate quality criterion were proved.