Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm
The prediction of imminent bankruptcy for a company is important to banks, government agencies, business owners, and different business stakeholders. Bankruptcy is influenced by many global and local aspects, so it can hardly be anticipated without deeper analysis and economic modeling knowledge. To...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2023-06-01
|
| Series: | PeerJ Computer Science |
| Subjects: | |
| Online Access: | https://peerj.com/articles/cs-1257.pdf |
| _version_ | 1797807038995103744 |
|---|---|
| author | Róbert Kanász Peter Gnip Martin Zoričák Peter Drotár |
| author_facet | Róbert Kanász Peter Gnip Martin Zoričák Peter Drotár |
| author_sort | Róbert Kanász |
| collection | DOAJ |
| description | The prediction of imminent bankruptcy for a company is important to banks, government agencies, business owners, and different business stakeholders. Bankruptcy is influenced by many global and local aspects, so it can hardly be anticipated without deeper analysis and economic modeling knowledge. To make this problem even more challenging, the available bankruptcy datasets are usually imbalanced since even in times of financial crisis, bankrupt companies constitute only a fraction of all operating businesses. In this article, we propose a novel bankruptcy prediction approach based on a shallow autoencoder ensemble that is optimized by a genetic algorithm. The goal of the autoencoders is to learn the distribution of the majority class: going concern businesses. Then, the bankrupt companies are represented by higher autoencoder reconstruction errors. The choice of the optimal threshold value for the reconstruction error, which is used to differentiate between bankrupt and nonbankrupt companies, is crucial and determines the final classification decision. In our approach, the threshold for each autoencoder is determined by a genetic algorithm. We evaluate the proposed method on four different datasets containing small and medium-sized enterprises. The results show that the autoencoder ensemble is able to identify bankrupt companies with geometric mean scores ranging from 71% to 93.7%, (depending on the industry and evaluation year). |
| first_indexed | 2024-03-13T06:16:31Z |
| format | Article |
| id | doaj.art-46bdf520ce0a4c3db015b4edbeecdafe |
| institution | Directory Open Access Journal |
| issn | 2376-5992 |
| language | English |
| last_indexed | 2024-03-13T06:16:31Z |
| publishDate | 2023-06-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ Computer Science |
| spelling | doaj.art-46bdf520ce0a4c3db015b4edbeecdafe2023-06-10T15:05:04ZengPeerJ Inc.PeerJ Computer Science2376-59922023-06-019e125710.7717/peerj-cs.1257Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithmRóbert Kanász0Peter Gnip1Martin Zoričák2Peter Drotár3Department of Computers and Informatics, Faculty of Electrical Engineering and Informatics, Technical University of Košice, Košice, SlovakiaDepartment of Computers and Informatics, Faculty of Electrical Engineering and Informatics, Technical University of Košice, Košice, SlovakiaDepartment of Finance, Faculty of Economics, Technical University of Košice, Košice, SlovakiaDepartment of Computers and Informatics, Faculty of Electrical Engineering and Informatics, Technical University of Košice, Košice, SlovakiaThe prediction of imminent bankruptcy for a company is important to banks, government agencies, business owners, and different business stakeholders. Bankruptcy is influenced by many global and local aspects, so it can hardly be anticipated without deeper analysis and economic modeling knowledge. To make this problem even more challenging, the available bankruptcy datasets are usually imbalanced since even in times of financial crisis, bankrupt companies constitute only a fraction of all operating businesses. In this article, we propose a novel bankruptcy prediction approach based on a shallow autoencoder ensemble that is optimized by a genetic algorithm. The goal of the autoencoders is to learn the distribution of the majority class: going concern businesses. Then, the bankrupt companies are represented by higher autoencoder reconstruction errors. The choice of the optimal threshold value for the reconstruction error, which is used to differentiate between bankrupt and nonbankrupt companies, is crucial and determines the final classification decision. In our approach, the threshold for each autoencoder is determined by a genetic algorithm. We evaluate the proposed method on four different datasets containing small and medium-sized enterprises. The results show that the autoencoder ensemble is able to identify bankrupt companies with geometric mean scores ranging from 71% to 93.7%, (depending on the industry and evaluation year).https://peerj.com/articles/cs-1257.pdfAutoencoderBankruptcy predictionImbalanced learningNeural networksGenetic algorithmFinancial ratios |
| spellingShingle | Róbert Kanász Peter Gnip Martin Zoričák Peter Drotár Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm PeerJ Computer Science Autoencoder Bankruptcy prediction Imbalanced learning Neural networks Genetic algorithm Financial ratios |
| title | Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm |
| title_full | Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm |
| title_fullStr | Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm |
| title_full_unstemmed | Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm |
| title_short | Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm |
| title_sort | bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm |
| topic | Autoencoder Bankruptcy prediction Imbalanced learning Neural networks Genetic algorithm Financial ratios |
| url | https://peerj.com/articles/cs-1257.pdf |
| work_keys_str_mv | AT robertkanasz bankruptcypredictionusingensembleofautoencodersoptimizedbygeneticalgorithm AT petergnip bankruptcypredictionusingensembleofautoencodersoptimizedbygeneticalgorithm AT martinzoricak bankruptcypredictionusingensembleofautoencodersoptimizedbygeneticalgorithm AT peterdrotar bankruptcypredictionusingensembleofautoencodersoptimizedbygeneticalgorithm |