Forecasting the bankruptcy of an organization based on metric methods of data mining
| Autor: | Lyudmila G. Matveeva, Igor V. Arinichev, Irina V. Arinicheva |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: | |
| Zdroj: | Voprosy Regulirovaniâ Èkonomiki, Vol 9, Iss 1, Pp 61-73 (2018) |
| ISSN: | 2412-6047 2078-5429 |
| Popis: | The article presents an approach to estimating the risk of bankruptcy of organizations of various fields of activity and industry, including water management complexes, based on inductive metric methods of data mining. One of the tasks of data analysis is the classification problem that arises in different spheres and branches of business. In the present paper a binary classification is considered when each object belongs to one of the two classes “bankrupt” or “not bankrupt”. A distinctive feature of this work is the use of qualitative factors - assessments of experts on the six features of this risk. The article compares five groups of methods with different types of distance functions, including the Euclidean metric and the Chebyshev distance, as well as four types of kernels for the method of potential functions. To adjust the parameters of the algorithms, a cross-validation of their quality is performed on the training and test samples. The simulation results showed that for some of the metrics in object space, the learning methods discussed show a good agreement with the initial data and demonstrate a small error on the test data. Training and optimization algorithms were implemented in the development environment of Visual Studio 2017 in the programming language C #. Given the simplicity of the implementation of metric methods, their reliability, the ability of algorithms to analyze significant amounts of information, it is assumed that the proposed approach to forecasting bankruptcy will be useful to representatives of small and medium-sized businesses and will provide an objective and accurate picture of the financial situation of the enterprise, current threats and the risk of bankruptcy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|