Information about imminent bankruptcy is crucial for financial institutions, decision-making managers, and state agencies. Since bankruptcy prediction is a prevalent research topic, many new methods have been continuously proposed. Bankruptcy prediction is frequently approached as a binary classification task. Since bankruptcy datasets are inherently imbalanced, bankruptcy classification is usually performed using class imbalance learning methods. The nature of these methods is very diverse, but they can usually be categorized as ensemble, cost-sensitive, sampling, and hybrid methods. In this paper, we provide a comprehensive experimental comparison of 45 methods. These methods were selected because they cover the approaches and algorithms frequently employed for bankruptcy prediction and imbalanced learning. Extensive experiments on 15 publicly available datasets with different imbalance ratios showed that the methods based on a combination of ensemble learning and undersampling are able to handle data imbalance and achieve the best results for bankruptcy classification.

Skewed data distribution poses many challenges in various domains, including the financial sector. Information about a company's potential bankruptcy is crucial for financial institutions and decision-making managers. As bankruptcy prediction has been a key concern for practitioners and research workers for decades, advancements in machine learning offer numerous methods to address imbalanced data scenarios with promising results. Among these, ensembles and neural networks showed remarkable results regarding prediction effectiveness. In this study, we propose a novel deep ensemble boosting approach to overcome imbalanced scenarios called Boosting TabNet. Achieved results showed promising results, specifically in the case of real-world datasets characterized by significant class imbalance ratios. Boosting TabNet outperformed or achieved results comparable to other approaches in most utilized datasets. The highest achieved score in terms of geometric mean (GM) score was over 96%.

In machine learning, particularly with tabular data, ensemble methods and neural networks stand as the preeminent approaches for predictive modeling. Among these, XGBoost and TabNet have demonstrated remarkable efficacy and interpretability. However, one critical challenge in these methodologies is their performance on imbalanced datasets, a common yet intricate issue in many real-world applications. This research paper proposes novel modifications to TabNet, tailored to enhance its performance on imbalanced tabular datasets. Our methodology introduces multiple loss functions for the TabNet architecture. These modifications improve the models’ sensitivity to minority classes and enhance overall predictive accuracy on imbalanced data. We conducted a comprehensive performance comparison using various synthetic and real-world datasets characterized by significant class imbalances. TabNet, combined with IBLoss, achieved a GM score on real-world data up to 92%. On synthetic data, the highest GM score was up to 82% using TabNet in combination with BVSLoss. These results demonstrate that TabNet is robust to imbalanced datasets and can learn well even on imbalanced data. The performance is further boosted by incorporating a loss function built for imbalanced data, such as BVSLoss or IBLoss. On the other hand, XGBoost fails to converge if not adapted to imbalanced data with sampling or cost-sensitive learning, resulting in less accurate prediction performance.

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).

Challenges posed by imbalanced data are encountered in many real-world applications. One of the possible approaches to improve the classifier performance on imbalanced data is oversampling. In this paper, we propose the new selective oversampling approach (SOA) that first isolates the most representative samples from minority classes by using an outlier detection technique and then utilizes these samples for synthetic oversampling. We show that the proposed approach improves the performance of two state-of-the-art oversampling methods, namely, the synthetic minority oversampling technique and adaptive synthetic sampling. The prediction performance is evaluated on four synthetic datasets and four real-world datasets, and the proposed SOA methods always achieved the same or better performance than other considered existing oversampling methods.

Bankruptcy prediction is still important topic receiving notable attention. Information about an imminent bankruptcy threat is a crucial aspect of the decision-making process of managers, financial institutions, and government agencies. In this paper, we utilize a newly acquired dataset comprising financial parameters derived from the annual reports of small- and medium-sized companies. The data, which reveal the true ratio between bankrupt and non-bankrupt companies, are severely imbalanced and only contain a small fraction of bankrupt companies. Our solution to overcome this challenging scenario of imbalanced learning was to adopt three one-class classification methods: a least-squares approach to anomaly detection, an isolation forest, and one-class support vector machines for comparison with conventional support vector machines. We provide a comprehensive analysis of the financial attributes and identify those that are most relevant to bankruptcy prediction. The highest prediction performance in terms of the geometric mean score is 91%. The results are validated on two datasets from the manufacturing and construction industries.

Application of the machine learning methods on strongly imbalanced datasets is a challenging task in the field of data processing. Imbalanced learning is part of many real-world applications and it is a very vivid research area. Moreover, bankruptcy prediction, even though it is one of the most popular prediction applications, is still not successfully solved. In this paper, we present a comparison of several ensemble machine learning methods applied on a recently acquired dataset of small and medium-sized enterprises operating in the Slovak Republic. The highest achieved prediction accuracy of the proposed classification models, measured by geometric mean, is in some cases almost 100%. Results are validated on three datasets from different business areas, namely agriculture, construction and retails.

Bankruptcy prediction is a long-standing issue that receives significant attention of academic researchers and industry practitioners. Most of the papers on bankruptcy prediction focus on companies that are listed on the stock market, and there are only limited data for the rest of the companies. These companies, not indexed at any stock market, represent a significant part of the economy. The presented dataset consists of financial ratios of Slovak companies. There are 21 distinctive financial ratios which are available for three consecutive years prior to evaluation year in which companies may have filed for bankruptcy or not. The companies come from four different industries - agriculture, construction, manufacture, retail. We provide data for four consecutive years 2013-2016 for each industry. All companies are categorized as small-medium enterprises according to EU classification. Prediction performance results on this dataset are published in the research paper "Bankruptcy prediction for small- and medium-sized companies using severely imbalanced datasets" (Zoricak et al., 2019). (c) 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Recently, huge amounts of data have been generated by computer and internet applications in multiple domains, including healthcare, bioinformatics, social media, e-commerce, and transportation. These data often have characteristics of high dimensions and their analysis is a challenge for researchers in the fields of machine learning and data mining. Feature selection is a dimensionality reduction technique that aims to select a subset of relevant features from the original feature set. Important aspect of feature selection, besides ability to identify all significant features, is the stability of the feature selection. In this paper, we investigate the stability of the Weighted k-Nearest Neighbors feature selection and compare it to other state-of-the art feature selection methods.

The companies involved in all areas of the business and industry can due to the unfavourable financial situation or inappropriate investments face financial problems resulting in bankruptcy of the company. The ability to foresee imminent bankruptcy helps managers and stock holders to take the corrective actions. In this paper, we analyze annual reports of thousands of limited liability companies and propose the bankruptcy prediction model. The available dataset is strongly imbalanced that corresponds to the real-world situation where bankrupt companies constitute only a small fraction of all companies. The proposed model is based on single-class least-squares anomaly detection classifier achieving as high as 91% prediction accuracy.