Staking approach to the problem of predicting bankruptcies

In this work, staking methods of predicting bankruptcy are improved and expanded. The main topic of this study is the use of different types of models: multilayer perceptrons, decision trees, Kohonen networks. In various combinations, they are selected and combined into ensembles and meta-ensembles in such a way as to consider the strengths and weaknesses of each type of basic models in the context of the task.