Systematic Review of Financial Distress Identification using Artificial Intelligence Methods

Figures & data

Figure 1. The procedure of the systematic review.

Table 1. Distribution of studies.

No.	Name of database	No. of studies	Ratio (%)
1	Science direct	85	36.6%
2	Springer	41	17.7%
3	Emerald	26	11.2%
4	IEEE	23	9.9%
5	ACM	10	4.3%
6	MDPI	10	4.3%
7	Ebsco	5	2.2%
8	Wiley	5	2.2%
9	Sage	3	1.3%
10	Taylor & Francis	2	0.9%
11	Other	22	9.5%
TOTAL:		232	100.0%

Figure 2. Domain research taxonomy.

Table 2. Distribution of data sources used in studies.

No.	Data source	Frequency	%
1	Stock exchange	69	36.3%
2	Polish	19	10.0%
3	Compustat	13	6.8%
4	Diane	13	6.8%
5	Private	10	5.3%
6	Taiwan Economic Journal (TEJ)	6	3.2%
7	Amadeus	5	2.6%
8	AIDA	4	2.1%
9	Japanese (JPNBDS)	3	1.6%
10	LoPucki’s Bankruptcy Research	3	1.6%
11	Other	45	23.7%
Total:		190	100.0%

Figure 3. Dimensionality reduction methods.

Figure 4. Class imbalance reduction methods.

Figure 5. Anomaly detection methods.

Table 3. Machine-learning methods applied in the context of financial distress.

No.	Method	Definition	Method modification	Source
Supervised machine-learning methods
1.	Aalen’s Additive Regression (AAR)	AAR is one of survival analysis Models (an alternative to Cox model). This model can detect changes in coefficient at each distinct survival time due to linear function in the hazard rate estimation.	$-$	(Zelenkov 2020)
2.	Accelerated Failure Time Models (AFT)	AFT is an alternative to proportional hazards models. AFT assumes that a covariate effect can become some constant, which accelerate or slow down the progression of a disease.	Weibull Accelerated Failure	(Zelenkov 2020)
			Time Models (WAF) (Zelenkov 2020)
3.	Artificial neural networks ANN Multilayer perceptron (MP|MLP)	The purpose of an artificial neural network (Multilayer perceptron) is to imitate the operation of biological neural systems, which is an example of neuron connectivity architecture design. The neurons are arranged in three completely connected layers: input, output, and one or more hidden layers. ANN provides a framework for modeling nonlinear functional mappings between sets of input and output variables. ANN with several hidden layers is also cold Deep neural network (DNN).	ANN-Backpropagation (Almaskati et al. 2021; Kristianto and Rikumahu 2019; Lin and Hsu 2017) Feed-forward neural network (FNN) (Adisa et al. 2019; AghaeiRad, Chen, and Ribeiro 2017; du Jardin 2018) FNN-SOM (AghaeiRad, Chen, and Ribeiro 2017) DGAMLPE (Zhang et al. 2021) DNN (Aljawazneh et al. 2021; Ben Jabeur, Stef, and Carmona 2022; Horak, Vrbka, and Suler 2020; Jabeur et al. 2021; Li et al. 2021; Smith and Alvarez 2022) GA-MLP (Zhang et al. 2021) MA-DNN (Li et al. 2021) Multi-task Neural Network (MNN) (Zelenkov 2020) Random vector functional-link network (RVFLN) (Lin and Hsu 2017) Rough set NN model (RSNN)(Acharjya and Rathi 2021) Self-organizing NN (du Jardin 2021a)	(Abdullah 2021; Acharjya and Rathi 2021; Adisa et al. 2019; AghaeiRad, Chen, and Ribeiro 2017; Aljawazneh et al. 2021; Aljawazneh et al. 2021; Almaskati et al. 2021; Azayite and Achchab 2018; Barboza, Kimura, and Altman 2017; Ben Jabeur, Stef, and Carmona 2022; Charalambous, Martzoukos, and Taoushianis 2022; Cheng, Chan, and Yang 2018; Choi, Son, and Kim 2018; Cleofas-Sánchez et al. 2016; Da et al. 2022; du Jardin 2018, 2021c; Fan, Liu, and Chen 2017; Faris et al. 2020; García, Marqués, and Sánchez 2019; Horak, Vrbka, and Suler 2020; Hosaka 2019; Hu 2020; Huang et al. 2017; Hájek 2018; Inam et al. 2019; Jabeur et al. 2021; Kim, Mun, and Bae 2018; Kristianto and Rikumahu 2019; Le et al. 2018, 2018; Letizia and Lillo 2019; Li et al. 2021; Liang et al. 2018; Lin and Hsu 2017; Mora García et al. 2008; Mousavi and Lin 2020; Mselmi, Lahiani, and Hamza 2017; Nyitrai and Virág 2019; Oz and Yelkenci 2017; Perboli and Arabnezhad 2021; Qian et al. 2022; Salehi and Davoudi Pour 2016; Salehi, Mousavi Shiri, and Bolandraftar Pasikhani 2016; Smith and Alvarez 2022; Smiti and Soui 2020; Soui et al. 2020; Sun et al. 2017; Séverin and Veganzones 2021; Tsai et al. 2021; Veganzones and Séverin 2018; Veganzones, Séverin, and Chlibi 2021; Vellamcheti and Singh 2020; Wagle, Yang, and Benslimane 2017; Ye 2021; Zelenkov 2020; Zelenkov, Fedorova, and Chekrizov 2017; Zhang et al. 2021; Zhou 2013)
4.	Bagging Bootstrap aggregation	An ensemble learning approach is a technique for reducing variance in a noisy data set. Bagging is the process of selecting a random sample of data from a training set with replacement—that is, the individual data points might be chosen many times. Another difference between the boosting and bagging techniques is that weak learners are trained parallel.	Bagging CART (Mousavi and Lin 2020) Balanced Bagging (Gnip and Drotár 2019) Random Subspace (du Jardin 2017, 2021a, 2021b, 2021c; Wang et al. 2020) IST-RS(Wang et al. 2018) Bagged-pSVM (Chen, Chen, and Shi 2020)	(Barboza, Kimura, and Altman 2017; Chen et al. 2021; Chen, Chen, and Shi 2020; du Jardin 2017, 2021a, 2021b, 2021c; Faris et al. 2020; Gnip and Drotár 2019; Keya et al. 2021; Liang et al. 2018; Mousavi and Lin 2020; Qian et al. 2022; Roumani, Nwankpa, and Tanniru 2020; Shen et al. 2020; Sisodia and Verma 2018; Sun et al. 2017; Wagle, Yang, and Benslimane 2017; Wang et al. 2020, 2018; Zelenkov and Volodarskiy 2021; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022)
5.	Boosting	An ensemble learning strategy for minimizing training errors by combining a group of weak learners into a strong learner. In boosting technique a group of weak learners are trained in a sequential manner.	AdaBoost (Aljawazneh et al. 2021; Faris et al. 2020; Hosaka 2019; Keya et al. 2021; Le et al. 2019; Qian et al. 2022; Shen et al. 2020; Sisodia and Verma 2018; Sun et al. 2017; Wang et al. 2020; Zelenkov and Volodarskiy 2021; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022); AdaBoost CART (Mousavi and Lin 2020); AdaBoost-SVM (Sun et al. 2017); ADASVM-TW (Sun et al. 2017, 2020); Boosted-pSVM (Chen, Chen, and Shi 2020); CatBoost (Jabeur et al. 2021; Zhao et al. 2022); CBoost (Le et al. 2019); CUSBoost (Du et al. 2020); CUS-GBDT (Du et al. 2020); Extreme Gradient Boosting (XGBoost) (Aljawazneh et al. 2021; Du et al. 2020; Huang and Yen 2019; Le et al. 2019; Perboli and Arabnezhad 2021; Qian et al. 2022; Shen et al. 2020; Sun et al. 2017; Zhao et al. 2022) (Ben Jabeur, Stef, and Carmona 2022; du Jardin 2021a, 2021b; c; Jabeur et al. 2021; Smith and Alvarez 2022; Son et al. 2019; Volkov, Benoit, and Van den Poel 2017); GMBoost (geometric mean) (Le et al. 2019); Gradient boosting model (GBoost) (Fan, Liu, and Chen 2017; Gnip and Drotár 2019; Jabeur et al. 2021; Jones 2017; Zelenkov and Volodarskiy 2021); LightGBM (Smith and Alvarez 2022; Son et al. 2019); Logit-Boost (Kumar and Roy 2016); NFS-Boost (Kumar and Roy 2016); RobustBoost (Pavlicko, Durica, and Mazanec 2021); RUSboost (Du et al. 2020; Gnip and Drotár 2019)	(Aljawazneh et al. 2021; Barboza, Kimura, and Altman 2017; Ben Jabeur, Stef, and Carmona 2022; Chen, Chen, and Shi 2020; Du et al. 2020; du Jardin 2017, 2021a, 2021b, 2021c; Fan, Liu, and Chen 2017; Faris et al. 2020; Gnip and Drotár 2019; Hosaka 2019; Huang and Yen 2019; Jabeur et al. 2021; Jones 2017; Keya et al. 2021; Kumar and Roy 2016; Le et al. 2018, 2019; Liang et al. 2018; Mousavi and Lin 2020; Pavlicko, Durica, and Mazanec 2021; Perboli and Arabnezhad 2021; Qian et al. 2022; Roumani, Nwankpa, and Tanniru 2020; Shen et al. 2020; Sisodia and Verma 2018; Smith and Alvarez 2022; Son et al. 2019; Soui et al. 2020; Sun et al. 2017, 2020; Vellamcheti and Singh 2020; Volkov, Benoit, and Van den Poel 2017; Wagle, Yang, and Benslimane 2017; Wang et al. 2020; Zelenkov and Volodarskiy 2021; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022)
6.	Cox: Cox’s model	Cox is essentially a regression model used for association understanding between time and one or more predictor variables, commonly used in medical research.	Cox proportional hazard model (Kanojia and Gupta 2022; Zelenkov 2020)	(du Jardin 2017, 2018; Kanojia and Gupta 2022; Roumani, Nwankpa, and Tanniru 2020; Zelenkov 2020)
7.	Data Envelopment Analysis (DEA)	DEA is a non-parametric approach, which empirically quantifies the relative evaluations of the efficiency between multiple similar factors inputs and outputs.	DEA - additive model (Almaskati et al. 2021)	(Almaskati et al. 2021)
8.	Decision tree (C4.5)	Decision tree consists of root, internal/decision node and leaf node. The model has to satisfy the following criteria: 1) The tree must have a single root, which has no entrance and from which the division into branches and leaves begins. 2) Each internal/decision node has only one input. 3) There is a unique path from each leaf node to the root. The splitting for internal/decision node is made by information gain ratio (C4.5), Gini coefficient (CART), ${\chi }^2$-based test (CHAID), etc. The decision tree pruning techniques are used to deal with overfitting. The decision tree is a nonparametric method, which means that it does not learn the parameters by which to evaluate attributes. The system remembers the main properties of the data. Therefore, even a small change in the data can lead to the formation of a new tree.	AdaBoost CART (Mousavi and Lin 2020); Bagging CART (Mousavi and Lin 2020); Bagging DT (Shen et al. 2020); C4.5 + Boosting (Putri and Dhini 2019); C4.5 + Bagging (Putri and Dhini 2019); C5.0 (Huang et al. 2017; Smiti and Soui 2020); CART (Behr and Weinblat 2017; Hosaka 2019; Jiang et al. 2021; Mousavi and Lin 2020; Nyitrai and Virág 2019; Pavlicko, Durica, and Mazanec 2021); CHAID (Nyitrai 2019; Nyitrai and Virág 2019); J48 (Chen et al. 2021; Keya et al. 2021; Vellamcheti and Singh 2020); JRip (Vellamcheti and Singh 2020); Logit regression tree (Andreano et al. 2018)	(Al-Milli, Hudaib, and Obeid 2021; Almaskati et al. 2021; Andreano et al. 2018; Behr and Weinblat 2017; Chen et al. 2021; Cheng, Chan, and Yang 2018; Choi, Son, and Kim 2018; du Jardin 2017, 2018; Faris et al. 2020; García, Marqués, and Sánchez 2019; Hosaka 2019; Hu 2020; Huang et al. 2017; Hájek 2018; Jiang et al. 2021; Keya et al. 2021; Kim and Upneja 2021; Le et al. 2018, 2018; Letizia and Lillo 2019; Liang et al. 2018; Mousavi and Lin 2020; Nyitrai 2019; Nyitrai and Virág 2019; Pavlicko, Durica, and Mazanec 2021; Putri and Dhini 2019; Qian et al. 2022; Roumani, Nwankpa, and Tanniru 2020; Shen et al. 2020; Sisodia and Verma 2018; Smiti and Soui 2020; Sun et al. 2017; Tsai et al. 2021; UlagaPriya and Pushpa 2021; Veganzones, Séverin, and Chlibi 2021; Vellamcheti and Singh 2020; Wagle, Yang, and Benslimane 2017; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022; Zhou 2013)
9.	Discriminant analysis (DA) Linear discriminant analysis (LDA) Multiple discriminant analysis (MDA)	The features combination usage for difference representation between groups, i.e. separating two or more classes. For example, LDA selects a new dimension that gives maximum separation between classes and minimum variance for each class.	PLS-DA (Ben Jabeur, Stef, and Carmona 2022; Mselmi, Lahiani, and Hamza 2017); Quadratic DA (Mousavi and Lin 2020; Zelenkov, Fedorova, and Chekrizov 2017)	(Al-Milli, Hudaib, and Obeid 2021; Almaskati et al. 2021; Altman 1968; Barboza, Kimura, and Altman 2017; Ben Jabeur, Stef, and Carmona 2022; du Jardin 2017, 2018; Hosaka 2019; Huang et al. 2017; Inam et al. 2019; Jabeur et al. 2021; Kim, Mun, and Bae 2018; Mousavi and Lin 2020; Mselmi, Lahiani, and Hamza 2017; Nyitrai 2019; Nyitrai and Virág 2019; Ptak-Chmielewska 2021; Soui et al. 2020; Séverin and Veganzones 2021; Veganzones and Séverin 2018; Vellamcheti and Singh 2020; Volkov, Benoit, and Van den Poel 2017; Zelenkov, Fedorova, and Chekrizov 2017; Zhou 2013)
10.	Gaussian processes (GP)	A stochastic process governs the properties of functions (Rasmussen and Williams 2006)	Multivariate Guassian (Fan, Liu, and Chen 2017); Normalized Gaussian RBF (Cleofas-Sánchez et al. 2016)	(Antunes, Ribeiro, and Pereira 2017; Cleofas-Sánchez et al. 2016; Fan, Liu, and Chen 2017)
11.	Genetic algorithm (GA)	The natural selection approach for optimization or search issues. Biological evolution has been driving force of GA, which consists of tree main operations: selection, crossover, mutation.	DGAMLPE (Zhang et al. 2021); GA-Fuzzy (Huang and Yen 2019); GA-MLP (Zhang et al. 2021); Genetic programming (GP) (Mora García et al. 2008); Rough set with genetic algorithm (Acharjya and Rathi 2021)	(Acharjya and Rathi 2021; Huang and Yen 2019; Mora García et al. 2008; Yang and Yang 2020)
12.	Hazard analysis (HA)	Hazard analysis consists of two time functions, which are called survival and default functions. The survival function shows the probability of survival beyond time (t), while function of default shows probability of failure at time (t), which means company has survived until that point in time.	$-$	(Alam, Gao, and Jones 2021; Almaskati et al. 2021; Escribano-Navas and Gemar 2021)
13.	k-Nearest neighbor k-NN	A non-parametric classifier that determines the class label for each unseen instance by a majority vote of its neighbors. The k is a number of nearest neighbors attended in the voting process.	Weightedvote Relational Neighbor (wvRN) (Tobback et al. 2017)	(Abdullah 2021; Al-Milli, Hudaib, and Obeid 2021; Aljawazneh et al. 2021; Chen et al. 2021; Choi, Son, and Kim 2018; Faris et al. 2020; García, Marqués, and Sánchez 2019; Hájek 2018; Jiang et al. 2021; Liang et al. 2018; Ouenniche, Pérez-Gladish, and Bouslah 2018; Pavlicko, Durica, and Mazanec 2021; Rahayu and Suhartanto 2020; Salehi, Mousavi Shiri, and Bolandraftar Pasikhani 2016; Smiti and Soui 2020; Tobback et al. 2017; Zelenkov, Fedorova, and Chekrizov 2017)
14.	Linear regression (LR) Multiple regression Multivariate regression	The linear approach for relationship modelling between dependent and independent variables. Simple or Multiple linear regression method usage depends on the number of independent variables used in the model (accordingly one or more than one).	OLS regression (Dumitrescu, El Hefnawy, and Zakriya 2020; Kovermann 2018)	(Aalbers et al. 2019; Boubaker et al. 2020; Dumitrescu, El Hefnawy, and Zakriya 2020; Kovermann 2018; Liu et al. 2021; Mathew, Ibrahim, and Archbold 2016; Ye 2021)
15.	Logistic regression (LR) Binary or Logit regression	A logistic function is used to model a binary dependent variable	Kernel ridge regression (KRR) (Shen et al. 2020); Logit regression tree (Andreano et al. 2018); Multinomial logistic regression (Hernandez Tinoco, Holmes, and Wilson 2018; Letizia and Lillo 2019); Panel LR (Ali et al. 2021; Oz and Yelkenci 2017; Succurro, Arcuri, and Costanzo 2019; Udin, Khan, and Javid 2017); Partial Least Squares LR (Ben Jabeur 2017); Probit model (Balasubramanian et al. 2019; Cooper and Uzun 2019; Oz and Simga-Mugan 2018)	(Ahmad 2019; Al-Milli, Hudaib, and Obeid 2021; Ali et al. 2021; Almaskati et al. 2021; Andreano et al. 2018; Antunes, Ribeiro, and Pereira 2017; Ashraf, Félix, and Serrasqueiro 2021; Awwad and Razia 2021; Balasubramanian et al. 2019; Barboza, Kimura, and Altman 2017; Behr and Weinblat 2017; Ben Jabeur 2017; Ben Jabeur, Stef, and Carmona 2022; Bravo-Urquiza and Moreno-Ureba 2021; Cenciarelli, Greco, and Allegrini 2018; Charalambous, Martzoukos, and Taoushianis 2022; Chen et al. 2021; Chiou, Lo, and Wu 2017; Choi, Son, and Kim 2018; Cleofas-Sánchez et al. 2016; Cooper and Uzun 2019; Crosato, Domenech, and Liberati 2021; Da et al. 2022; Darrat et al. 2016; Doğan, Koçak, and Atan 2022; du Jardin 2017, 2018; Freitas Cardoso, Peixoto, and Barboza 2019; García and Herrero 2021; Hernandez Tinoco, Holmes, and Wilson 2018; Hosaka 2019; Hu 2020; Huang et al. 2017, 2017; Inam et al. 2019; Jabeur et al. 2021; Jiang et al. 2021; Jindal 2020; Kamalirezaei et al. 2020; Kanojia and Gupta 2022; Kim, Cho, and Ryu 2021; Kim, Mun, and Bae 2018; Letizia and Lillo 2019; Li and Faff 2019; Li et al. 2021; Lin and Dong 2018; Mai et al. 2019; Mousavi and Lin 2020; Mselmi, Lahiani, and Hamza 2017; Nyitrai 2019; Nyitrai and Virág 2019; Olsen and Tamm 2017; Oz and Yelkenci 2017; Park et al. 2021; Perboli and Arabnezhad 2021; Pham Vo Ninh, Do Thanh, and Vo Hong 2018; Putri and Dhini 2019; Qian et al. 2022; Rahayu and Suhartanto 2020; Ravula 2021; Regenburg and Seitz 2021; Ren, Lu, and Yang 2021; Richardson, Lanis, and Taylor 2015; Roumani, Nwankpa, and Tanniru 2020; Sayari and Mugan 2017; Shahwan and Habib 2020; Shen et al. 2020; Sisodia and Verma 2018; Smith and Alvarez 2022; Son et al. 2019; Soui et al. 2020; Stef and Zenou 2021; Succurro, Arcuri, and Costanzo 2019; Sun et al. 2017; Séverin and Veganzones 2021; Süsi and Lukason 2019; Tsai et al. 2021; Udin, Khan, and Javid 2017; Veganzones and Séverin 2018; Veganzones, Séverin, and Chlibi 2021; Volkov, Benoit, and Van den Poel 2017; Wagle, Yang, and Benslimane 2017; Yazdanfar and Öhman 2020; Zelenkov and Volodarskiy 2021; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022; Zhou 2013)
16.	Naive Bayes	Classification algorithm that is based on the statistical Bayes theorem, along with the assumption of feature independence. Naive Bayes determines the instance class by the probability calculation for the instance belonging in the certain class.	Gaussian Naive Bayes (Rahayu and Suhartanto 2020)	(Abdullah 2021; Chen et al. 2021; Choi, Son, and Kim 2018; Cleofas-Sánchez et al. 2016; Faris et al. 2020; Hájek 2018; Liang et al. 2018; Mousavi and Lin 2020; Rahayu and Suhartanto 2020; Roumani, Nwankpa, and Tanniru 2020; Salehi, Mousavi Shiri, and Bolandraftar Pasikhani 2016; Shen et al. 2020; Wagle, Yang, and Benslimane 2017; Zelenkov, Fedorova, and Chekrizov 2017)
17.	Random Forest (RF)	The type of ensemble estimator method which consist of creating multi-decision trees using various samples from the original data set. Each tree in RF is generated from a bootstrap sample of the data, and a random sample of predictors is inspected at each split. The classification class is determined by the majority voting from the decision trees.	Balanced Random Forest (Gnip and Drotár 2019); Oblique random forests (obRF) (Shen et al. 2020); Random Survival Forest (RSF) (Zelenkov 2020); Rotation forest (du Jardin 2017, 2021b, 2021c; Faris et al. 2020)	(Aljawazneh et al. 2021; Angenent, Barata, and Takes 2020; Barboza, Kimura, and Altman 2017; Behr and Weinblat 2017; Chang 2019; Da et al. 2022; du Jardin 2017, 2021a, 2021b, 2021c; Fan, Liu, and Chen 2017; Faris et al. 2020; Gnip and Drotár 2019; Huang et al. 2017; Jabeur et al. 2021; Jiang et al. 2021; Jindal 2020; Joshi, Ramesh, and Tahsildar 2018; Keya et al. 2021; Kim, Cho, and Ryu 2021; Le et al. 2018, 2018; Mai et al. 2019; Mousavi and Lin 2020; Perboli and Arabnezhad 2021; Qian et al. 2022; Shen et al. 2020; Sisodia and Verma 2018; Smith and Alvarez 2022; Smiti and Soui 2020; Son et al. 2019; Soui et al. 2020; Sun et al. 2017; Veganzones and Séverin 2018; Vellamcheti and Singh 2020; Volkov, Benoit, and Van den Poel 2017; Ye 2021; Zelenkov 2020; Zelenkov and Volodarskiy 2021; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022)
18.	Support vector machine (SVM)	SVM uses high-dimensional feature space for class separation by the identification of optimal separating hyperplane (decision boundary line). SVM is applied for linearly (separable by a linear hyperplane) and non-linearly separable data. The SVM approach uses quadratic programming to find unique solutions to the idea of structural risk reduction, which tries to lower the boundaries of misclassification errors by generating an optimal separating hyperplane in a high-dimensional feature space.	AdaBoost-SVM (Sun et al. 2017); ADASVM-TW (Sun et al. 2017, 2020); Bagged-pSVM (Chen, Chen, and Shi 2020); Boosted-pSVM (Chen, Chen, and Shi 2020); DBN$-$SVM (Huang and Yen 2019); DD-SVM (Kim, Mun, and Bae 2018); OS-SVM (Wang et al. 2020); PLS-SVM (Mselmi, Lahiani, and Hamza 2017); Relevance vector machine (RVM) (Lin and Hsu 2017); SMOTE-SVM (Wang et al. 2020); Timeboost-SVM (Sun et al. 2017); US-SVM (Wang et al. 2020)	(Abdullah 2021; Al-Milli, Hudaib, and Obeid 2021; Aljawazneh et al. 2021; Antunes, Ribeiro, and Pereira 2017; Barboza, Kimura, and Altman 2017; Ben Jabeur, Stef, and Carmona 2022; Chang 2019; Chen, Chen, and Shi 2020; Cheng, Chan, and Yang 2018; Choi, Son, and Kim 2018; Cleofas-Sánchez et al. 2016; Da et al. 2022; Doğan, Koçak, and Atan 2022; du Jardin 2017, 2018; Horak, Vrbka, and Suler 2020; Hosaka 2019; Hu 2020; Huang and Yen 2019; Huang et al. 2017, 2017; Hájek 2018; Jabeur et al. 2021; Kim, Cho, and Ryu 2021; Kim, Mun, and Bae 2018; Le et al. 2018; Liang et al. 2018; Lin and Hsu 2017; Lu et al. 2014; Mai et al. 2019; Mousavi and Lin 2020; Mselmi, Lahiani, and Hamza 2017; Ptak-Chmielewska 2021; Qian et al. 2022; Rahayu and Suhartanto 2020; Ren, Lu, and Yang 2021; Salehi, Mousavi Shiri, and Bolandraftar Pasikhani 2016; Shen et al. 2020; Sisodia and Verma 2018; Smith and Alvarez 2022; Smiti and Soui 2020; Soui et al. 2020; Sun et al. 2017, 2020, 2021; Séverin and Veganzones 2021; Tobback et al. 2017; Tsai and Wang 2017; Tsai et al. 2021; Veganzones and Séverin 2018; Veganzones, Séverin, and Chlibi 2021; Vellamcheti and Singh 2020; Volkov, Benoit, and Van den Poel 2017; Wagle, Yang, and Benslimane 2017; Wang et al. 2020; Ye 2021; Zelenkov, Fedorova, and Chekrizov 2017; Zhao et al. 2022; Zhou 2013; Zoričák et al. 2020)
19.	TOPSIS	Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) provides ranking alternatives scores, which is known as the positive and the negative ideal solutions, and is based on geometric distance calculation.	-	(Ouenniche, Pérez-Gladish, and Bouslah 2018)
20.	VIKOR	VIKOR is a multi-criteria method created for the ranking a variety of options. VIKOR is designed to chose the compromise solution that is closest to the ideal.	-	(Ouenniche et al. 2021)
Deep learning models
21.	Convolutional neural network (ConvNet|CNN)	CNN is deep learning architecture that learns directly from data, without additional need for feature extraction. CNN architecture consists of convolution, pooling, and fully connected layers, which are used for automatic and adoptive learn of features for the classification tasks.	Deep Grassmannian Network (GrNet) (Alam, Gao, and Jones 2021); Dependency Sensitive Convolutional Neural Networks (DSCNNs) (Ahmadi et al. 2018)	(Abdi et al. 2019; Ahmadi et al. 2018; Alam, Gao, and Jones 2021; Hosaka 2019; Mai et al. 2019; Xu, Fu, and Pan 2019)
22.	Deep Belief Network (DBN)	The strength of DBN is its capacity to recreate both the input and learning feature vectors using a layer-by-layer learning technique. Each layer of a DBN consists of Restricted Boltzmann Machines (RBM). The neurons between different layers are fully connected, and the neurons in the same layer are not connected.	-	(Aljawazneh et al. 2021)
23.	Deep Boltzmann machines (DBM)	DBM is comprised of Restricted Boltzmann Machines (RBM). The DBM units in the same layer have no direct connections. The estimated gradient of a variational lower bound on the likelihood goal can be used to optimize the DBM parameters of all layers at the same time.	DBN-SVM (Huang and Yen 2019)	(Huang and Yen 2019)
24.	Extreme learning machine (ELM)	The extreme learning machine (ELM) is a single hidden layer feedforward networks.The weights between the input and hidden layers are assigned at random using a normal distribution, and the weights between the hidden and output layers are learned in a single step using a pseudo-inverse technique in ELM. During the training, the output weights are solved from regularized least squares.	GA-KELM (Wang et al. 2017); GS-KELM (Wang et al. 2017); GWO-KELM (Wang et al. 2017); Kernel extreme learning machine (KELM) (Zhang et al. 2021); PSO-KELM (Wang et al. 2017)	(du Jardin 2017, 2018; Séverin and Veganzones 2021; Veganzones, Séverin, and Chlibi 2021; Zhang et al. 2021)
25.	Long Short-Term Memory (LSTM)	LSTM advanced or extension of RNN. LSTM processes information in a sequential manner, but there is a memory cell that remembers and forgets information. Three multiplication units regulate the flow of information in each memory cell: input gate, output gate, and forget gate. The three gates control the flow of information into and out of the cell, and the cell recalls values across arbitrary timestamps.	Bi-LSMT (Li et al. 2021); Dependency Sensitive RNSA (Abdi et al. 2019)	(Abdi et al. 2019; Aljawazneh et al. 2021; Da et al. 2022; Kim, Cho, and Ryu 2021)
26.	Recurrent Neural Network	An agent learns from interactions with its environment in discrete time steps to update its mapping between the observed state and the probability of selecting possible actions in a reinforcement learning process. Q-learning, as a typical reinforcement learning approach, imitates human behavior by taking actions in the environment to maximize long-term rewards.		(Abdi et al. 2019; Kim, Cho, and Ryu 2021)
27.	Recursive deep learning	An RNN has feedback connections, which can be between hidden units or from the output to the hidden units. It is able to process the sequential inputs by having a recurrent hidden state whose activation at each step depends on that of the previous steps		(Abdi et al. 2019)
Unsupervised machine-learning methods
1.	K-means	K-means aggregates data points based on certain similarities (nearest mean, cluster centers or centroid) into k clusters.	Fuzzy C-means (Chou, Hsieh, and Qiu 2017)	(Chou, Hsieh, and Qiu 2017)
2.	Isolation Forest (IF)	Isolation Forest assumes that anomaly has two quantitative features: 1) a small minority of examples, and 2) their feature-values are different from a normal class examples. Isolation Forest applies iTree (binary tree) for effective splitting and isolating examples. The anomaly scores generated from the path length of each observed example from root node to node in which the example is isolated are used to create the final model.	-	(Fan, Liu, and Chen 2017; Gnip and Drotár 2019; Zoričák et al. 2020)
3.	Least-Squares Anomaly Detection (LSAD)	LSAD is based on the idea of One-Class SVM, which uses a hypersphere to encompass all the instances. Though, LSAD is used an extended application of the least squares probabilistic classification for a lost function.		(Zoričák et al. 2020)
4.	One-Class SVM (OCSVM)	OCSVM aim is to make separation between majority (training) class instances and outliers instances, which are points laying outside captured characteristics of training instances.	-	(Fan, Liu, and Chen 2017; Gnip and Drotár 2019; Zoričák et al. 2020)
Deep learning models
5.	Auto-encoder	An Autoencoder is a neural network type, where the input and the output are the same.The autoencoder consists of: input layer, one or more hidden layers (encoding layers), and output layer (decoding layer). The autoencoder’s primary goal is to lower the dimensionality of the input data by reducing noise.	Stacked Auto-encoder (Soui et al. 2020)	(Smiti and Soui 2020; Soui et al. 2020)

Figure 6. Quantity and number of studies of evaluation metrics used in studies.

Zelenkov, Y. (2020). Bankruptcy prediction using survival analysis technique. In 2020 IEEE 22nd Conference on Business Informatics (CBI), Antwerp, Belgium, Volume 2, pp. 141–49.

 Google Scholar

Almaskati, N., R. Bird, D. Yeung, and Y. Lu. 2021. A horse race of models and estimation methods for predicting bankruptcy. Advances in Accounting 52:100513. doi:10.1016/j.adiac.2021.100513.

 Web of Science ®Google Scholar

Kristianto, H., and B. Rikumahu. (2019). A cross model telco industry financial distress prediction in Indonesia: multiple discriminant analysis, Logit and Artificial Neural Network. In 2019 7th International Conference on Information and Communication Technology (ICoICT), Kuala Lumpur, Malaysia, pp. 1–5.

 Google Scholar

Lin, S.-J., and M.-F. Hsu. 2017. Incorporated risk metrics and hybrid AI techniques for risk management. Neural Computing & Applications 28 (11):3477–89. doi:10.1007/s00521-016-2253-4.

 Web of Science ®Google Scholar

Adisa, J. A., S. O. Ojo, P. A. Owolawi, and A. B. Pretorius. (2019). Financial distress prediction: principle component analysis and artificial neural networks. In 2019 International Multidisciplinary Information Technology and Engineering Conference (IMITEC), Vanderbijlpark, South Africa, pp. 1–6.

 Google Scholar

AghaeiRad, A., N. Chen, and B. Ribeiro. 2017. Improve credit scoring using transfer of learned knowledge from self-organizing map. Neural Computing & Applications 28 (6):1329–42. doi:10.1007/s00521-016-2567-2.

 Web of Science ®Google Scholar

du Jardin, P. 2018. Failure pattern-based ensembles applied to bankruptcy forecasting. Decision Support Systems 107:64–77. doi:10.1016/j.dss.2018.01.003.

 Web of Science ®Google Scholar

Zhang, R., Z. Zhang, D. Wang, and M. Du. 2021. Financial distress prediction with a novel diversity-considered GA-MLP ensemble algorithm. Neural Processing Letters 54 (2):1175–94. doi:10.1007/s11063-021-10674-9.

 Web of Science ®Google Scholar

Aljawazneh, H., A. M. Mora, P. Garcia-Sanchez, and P. A. Castillo-Valdivieso. 2021. Comparing the performance of deep learning methods to predict companies’ financial failure. IEEE Access 9:97010–38. doi:10.1109/ACCESS.2021.3093461.

 Web of Science ®Google Scholar

Ben Jabeur, S., N. Stef, and P. Carmona. 2022 Bankruptcy prediction using the XGBoost algorithm and variable importance feature engineering . Computational Economics. doi:10.1007/s10614-021-10227-1.

 Web of Science ®Google Scholar

Horak, J., J. Vrbka, and P. Suler. 2020. Support vector machine methods and artificial neural networks used for the development of bankruptcy prediction models and their comparison. Journal of Risk and Financial Management 13 (3):60. doi:10.3390/jrfm13030060.

 Web of Science ®Google Scholar

Jabeur, S. B., C. Gharib, S. Mefteh-Wali, and W. B. Arfi. 2021. CatBoost model and artificial intelligence techniques for corporate failure prediction. Technological Forecasting and Social Change 166:120658. doi:10.1016/j.techfore.2021.120658.

 Web of Science ®Google Scholar

Li, C., C. Lou, D. Luo, and K. Xing. 2021. Chinese corporate distress prediction using LASSO: The role of earnings management. International Review of Financial Analysis 76:101776. doi:10.1016/j.irfa.2021.101776.

 Web of Science ®Google Scholar

Smith, M., and F. Alvarez. 2022. Predicting firm-level bankruptcy in the Spanish economy using extreme gradient boosting. Computational Economics 59 (1):263–95. doi:10.1007/s10614-020-10078-2.

 Web of Science ®Google Scholar

Li, S., W. Shi, J. Wang, and H. Zhou. 2021. A deep learning-based approach to constructing a domain sentiment lexicon: a case study in financial distress prediction. Information Processing & Management 58 (5):102673. doi:10.1016/j.ipm.2021.102673.

 Web of Science ®Google Scholar

Acharjya, D. P., and R. Rathi. 2021. An extensive study of statistical, rough, and hybridized rough computing in bankruptcy prediction. Multimedia Tools and Applications 80 (28–29):35387–413. doi:10.1007/s11042-020-10167-2.

 Web of Science ®Google Scholar

du Jardin, P. 2021a. Dynamic self-organizing feature map-based models applied to bankruptcy prediction. Decision Support Systems 147:113576. doi:10.1016/j.dss.2021.113576.

 Web of Science ®Google Scholar

Abdullah, M. 2021. The implication of machine learning for financial solvency prediction: An empirical analysis on public listed companies of Bangladesh. Journal of Asian Business and Economic Studies 28 (4):303–20. doi:10.1108/JABES-11-2020-0128.

 Google Scholar

Azayite, F. Z., and S. Achchab. (2018). Topology design of bankruptcy prediction neural networks using Particle swarm optimization and backpropagation. In Proceedings of the International Conference on Learning and Optimization Algorithms: Theory and Applications, LOPAL ’18, New York, NY, USA, pp. 1–6.

 Google Scholar

Barboza, F., H. Kimura, and E. Altman. 2017. Machine learning models and bankruptcy prediction. Expert Systems with Applications 83:405–17. doi:10.1016/j.eswa.2017.04.006.

 Web of Science ®Google Scholar

Charalambous, C., S. H. Martzoukos, and Z. Taoushianis. 2022. Estimating corporate bankruptcy forecasting models by maximizing discriminatory power. Review of Quantitative Finance and Accounting 58 (1):297–328. doi:10.1007/s11156-021-00995-0.

 Web of Science ®Google Scholar

Cheng, C.-H., C.-P. Chan, and J.-H. Yang. 2018. A seasonal time-series model based on gene expression programming for predicting financial distress. Computational Intelligence & Neuroscience 2018:1–14. doi: 10.1155/2018/1067350

 Web of Science ®Google Scholar

Choi, H., H. Son, and C. Kim. 2018. Predicting financial distress of contractors in the construction industry using ensemble learning. Expert Systems with Applications 110:1–10. doi:10.1016/j.eswa.2018.05.026.

 Web of Science ®Google Scholar

Cleofas-Sánchez, L., V. García, A. Marqués, and J. Sánchez. 2016. Financial distress prediction using the hybrid associative memory with translation. Applied Soft Computing 44:144–52. doi:10.1016/j.asoc.2016.04.005.

 Web of Science ®Google Scholar

Da, F., C. Peng, H. Wang, and T. Li. 2022. A risk detection framework of Chinese high-tech firms using wide & deep learning model based on text disclosure. Procedia computer science 199:262–68. doi:10.1016/j.procs.2022.01.032.

 Google Scholar

du Jardin, P. 2021c. Forecasting corporate failure using ensemble of self-organizing neural networks. European Journal of Operational Research 288 (3):869–85. doi:10.1016/j.ejor.2020.06.020.

 Web of Science ®Google Scholar

Fan, S., G. Liu, and Z. Chen. (2017). Anomaly detection methods for bankruptcy prediction. In 2017 4th International Conference on Systems and Informatics (ICSAI), Hangzhou, China, pp. 1456–60.

 Google Scholar

Faris, H., R. Abukhurma, W. Almanaseer, M. Saadeh, A. M. Mora, P. A. Castillo, and I. Aljarah. 2020. Improving financial bankruptcy prediction in a highly imbalanced class distribution using oversampling and ensemble learning: A case from the Spanish market. Progress in Artificial Intelligence 9 (1):31–53. doi:10.1007/s13748-019-00197-9.

 Web of Science ®Google Scholar

García, V., A. I. Marqués, and J. S. Sánchez. 2019. Exploring the synergetic effects of sample types on the performance of ensembles for credit risk and corporate bankruptcy prediction. Information Fusion 47:88–101. doi:10.1016/j.inffus.2018.07.004.

 Web of Science ®Google Scholar

Hosaka, T. 2019. Bankruptcy prediction using imaged financial ratios and convolutional neural networks. Expert Systems with Applications 117:287–99. doi:10.1016/j.eswa.2018.09.039.

 Web of Science ®Google Scholar

Hu, Y.-C. 2020. A multivariate grey prediction model with grey relational analysis for bankruptcy prediction problems. Soft Computing 24 (6):4259–68. doi:10.1007/s00500-019-04191-0.

 Web of Science ®Google Scholar

Huang, J., H. Wang, and G. Kochenberger. 2017. Distressed Chinese firm prediction with discretized data. Management Decision 55 (5):786–807. doi:10.1108/MD-08-2016-0546.

 Web of Science ®Google Scholar

Hájek, P. 2018. Combining bag-of-words and sentiment features of annual reports to predict abnormal stock returns. Neural Computing & Applications 29 (7):343–58. doi:10.1007/s00521-017-3194-2.

 Web of Science ®Google Scholar

Inam, F., A. Inam, M. A. Mian, A. A. Sheikh, and H. M. Awan. 2019. Forecasting Bankruptcy for organizational sustainability in Pakistan: Using artificial neural networks, logit regression, and discriminant analysis. Journal of Economic and Administrative Sciences 35 (3):183–201. doi:10.1108/JEAS-05-2018-0063.

 Web of Science ®Google Scholar

Kim, S., B. M. Mun, and S. J. Bae. 2018. Data depth based support vector machines for predicting corporate bankruptcy. Applied Intelligence 48 (3):791–804. doi:10.1007/s10489-017-1011-3.

 Web of Science ®Google Scholar

Le, T., L. Hoang Son, M. Vo, M. Lee, and S. Baik. 2018. A cluster-based boosting algorithm for bankruptcy prediction in a highly imbalanced dataset. Symmetry 10 (7):250. doi:10.3390/sym10070250.

 Google Scholar

Le, T., M. Lee, J. Park, and S. Baik. 2018. Oversampling techniques for bankruptcy prediction: novel features from a transaction dataset. Symmetry 10 (4):79. doi:10.3390/sym10040079.

 Google Scholar

Letizia, E., and F. Lillo. 2019. Corporate payments networks and credit risk rating. EPJ Data Science 8 (1):21. doi:10.1140/epjds/s13688-019-0197-5.

 Google Scholar

Liang, D., C.-F. Tsai, A.-J. Dai, and W. Eberle. 2018. A novel classifier ensemble approach for financial distress prediction. Knowledge and Information Systems 54 (2):437–62. doi:10.1007/s10115-017-1061-1.

 Web of Science ®Google Scholar

Mora García, A. M., P. A. Castillo Valdivieso, J. J. Merelo Guervós, E. Alfaro Cid, A. I. Esparcia-Alcázar, and K. Sharman. (2008). Discovering causes of financial distress by combining evolutionary algorithms and artificial neural networks. In Proceedings of the 10th annual conference on Genetic and evolutionary computation, GECCO ’08, New York, NY, USA, pp. 1243–50.

 Google Scholar

Mousavi, M. M., and J. Lin. 2020. The application of PROMETHEE multi-criteria decision aid in financial decision making: Case of distress prediction models evaluation. Expert Systems with Applications 159:113438. doi:10.1016/j.eswa.2020.113438.

 Web of Science ®Google Scholar

Mselmi, N., A. Lahiani, and T. Hamza. 2017. Financial distress prediction: The case of French small and medium-sized firms. International Review of Financial Analysis 50:67–80. doi:10.1016/j.irfa.2017.02.004.

 Web of Science ®Google Scholar

Nyitrai, T., and M. Virág. 2019. The effects of handling outliers on the performance of bankruptcy prediction models. Socio-Economic Planning Sciences 67:34–42. doi:10.1016/j.seps.2018.08.004.

 Web of Science ®Google Scholar

Oz, I. O., and T. Yelkenci. 2017. A theoretical approach to financial distress prediction modeling. Managerial Finance 43 (2):212–30. doi:10.1108/MF-03-2016-0084.

 Web of Science ®Google Scholar

Perboli, G., and E. Arabnezhad. 2021. A Machine Learning-based DSS for mid and long-term company crisis prediction. Expert Systems with Applications 174:114758. doi:10.1016/j.eswa.2021.114758.

 Web of Science ®Google Scholar

Qian, H., B. Wang, M. Yuan, S. Gao, and Y. Song. 2022. Financial distress prediction using a corrected feature selection measure and gradient boosted decision tree. Expert Systems with Applications 190:116202. doi:10.1016/j.eswa.2021.116202.

 Web of Science ®Google Scholar

Salehi, M., and M. Davoudi Pour. 2016. Bankruptcy prediction of listed companies on the Tehran Stock Exchange. International Journal of Law and Management 58 (5):545–61. doi:10.1108/IJLMA-05-2015-0023.

 Google Scholar

Salehi, M., M. Mousavi Shiri, and M. Bolandraftar Pasikhani. 2016. Predicting corporate financial distress using data mining techniques: An application in Tehran Stock Exchange. International Journal of Law and Management 58 (2):216–30. doi:10.1108/IJLMA-06-2015-0028.

 Google Scholar

Smiti, S., and M. Soui. 2020. Bankruptcy prediction using deep learning approach based on borderline SMOTE. Information Systems Frontiers 22 (5):1067–83. doi:10.1007/s10796-020-10031-6.

 Web of Science ®Google Scholar

Soui, M., S. Smiti, M. W. Mkaouer, and R. Ejbali. 2020. Bankruptcy prediction using stacked auto-encoders. Applied Artificial Intelligence 34 (1):80–100. doi:10.1080/08839514.2019.1691849.

 Web of Science ®Google Scholar

Sun, J., H. Fujita, P. Chen, and H. Li. 2017. Dynamic financial distress prediction with concept drift based on time weighting combined with Adaboost support vector machine ensemble. Knowledge-Based Systems 120:4–14. doi:10.1016/j.knosys.2016.12.019.

 Web of Science ®Google Scholar

Séverin, E., and D. Veganzones. 2021. Can earnings management information improve bankruptcy prediction models? Annals of Operations Research 306 (1–2):247–72. doi:10.1007/s10479-021-04183-0.

 Web of Science ®Google Scholar

Tsai, C.-F., K.-L. Sue, Y.-H. Hu, and A. Chiu. 2021. Combining feature selection, instance selection, and ensemble classification techniques for improved financial distress prediction. Journal of Business Research 130:200–09. doi:10.1016/j.jbusres.2021.03.018.

 Web of Science ®Google Scholar

Veganzones, D., and E. Séverin. 2018. An investigation of bankruptcy prediction in imbalanced datasets. Decision Support Systems 112:111–24. doi:10.1016/j.dss.2018.06.011.

 Web of Science ®Google Scholar

Veganzones, D., E. Séverin, and S. Chlibi. 2021. Influence of earnings management on forecasting corporate failure. International Journal of Forecasting S0169207021001503. doi:10.1016/j.ijforecast.2021.09.006.

 Web of Science ®Google Scholar

Vellamcheti, S., and P. Singh. (2020). Class imbalance deep learning for bankruptcy prediction. In 2020 First International Conference on Power, Control and Computing Technologies (ICPC2T), Raipur, India, pp. 421–25.

 Google Scholar

Wagle, M., Z. Yang, and Y. Benslimane. (2017). Bankruptcy prediction using data mining techniques. In 2017 8th International Conference of Information and Communication Technology for Embedded Systems (IC-ICTES), Chonburi , Thailand, pp. 1–4.

 Google Scholar

Ye, Z. (2021). A data slicing method to improve machine learning model accuracy in bankruptcy prediction. 2021 5th International Conference on Deep Learning Technologies (ICDLT), Qingdao China, pp. 32–39.

 Google Scholar

Zelenkov, Y., E. Fedorova, and D. Chekrizov. 2017. Two-step classification method based on genetic algorithm for bankruptcy forecasting. Expert Systems with Applications 88:393–401. doi:10.1016/j.eswa.2017.07.025.

 Web of Science ®Google Scholar

Zhou, L. 2013. Performance of corporate bankruptcy prediction models on imbalanced dataset: The effect of sampling methods. Knowledge-Based Systems 41:16–25. doi:10.1016/j.knosys.2012.12.007.

 Web of Science ®Google Scholar

Gnip, P., and P. Drotár. (2019). Ensemble methods for strongly imbalanced data: Bankruptcy prediction. In 2019 IEEE 17th International Symposium on Intelligent Systems and Informatics (SISY), Subotica, Serbia, pp. 155–60.

 Google Scholar

du Jardin, P. 2017. Dynamics of firm financial evolution and bankruptcy prediction. Expert Systems with Applications 75:25–43. doi:10.1016/j.eswa.2017.01.016.

 Web of Science ®Google Scholar

du Jardin, P. 2021b. Forecasting bankruptcy using biclustering and neural network-based ensembles. Annals of Operations Research 299 (1–2):531–66. doi:10.1007/s10479-019-03283-2.

 Web of Science ®Google Scholar

Wang, G., J. Ma, G. Chen, and Y. Yang. 2020. Financial distress prediction: Regularized sparse-based Random Subspace with ER aggregation rule incorporating textual disclosures. Applied Soft Computing 90:106152. doi:10.1016/j.asoc.2020.106152.

 Web of Science ®Google Scholar

Wang, G., G. Chen, and Y. Chu. 2018. A new random subspace method incorporating sentiment and textual information for financial distress prediction. Electronic Commerce Research and Applications 29:30–49. doi:10.1016/j.elerap.2018.03.004.

 Web of Science ®Google Scholar

Chen, Z., W. Chen, and Y. Shi. 2020. Ensemble learning with label proportions for bankruptcy prediction. Expert Systems with Applications 146:113155. doi:10.1016/j.eswa.2019.113155.

 Web of Science ®Google Scholar

Chen, Y.-S., C.-K. Lin, C.-M. Lo, S.-F. Chen, and Q.-J. Liao. 2021. Comparable studies of financial bankruptcy prediction using advanced hybrid intelligent classification models to provide early warning in the electronics industry. Mathematics 9 (20):2622. doi:10.3390/math9202622.

 Web of Science ®Google Scholar

Keya, M. S., H. Akter, M. A. Rahman, M. M. Rahman, M. U. Emon, and M. S. Zulfiker. (2021). Comparison of different machine learning algorithms for detecting bankruptcy. In 2021 6th International Conference on Inventive Computation Technologies (ICICT), Coimbatore, India, pp. 705–12. IEEE.

 Google Scholar

Roumani, Y. F., J. K. Nwankpa, and M. Tanniru. 2020. Predicting firm failure in the software industry. Artificial Intelligence Review 53 (6):4161–82. doi:10.1007/s10462-019-09789-2.

 Web of Science ®Google Scholar

Shen, F., Y. Liu, R. Wang, and W. Zhou. 2020. A dynamic financial distress forecast model with multiple forecast results under unbalanced data environment. Knowledge-Based Systems 192:105365. doi:10.1016/j.knosys.2019.105365.

 Web of Science ®Google Scholar

Sisodia, D. S., and U. Verma. 2018. The impact of data re-sampling on learning performance of class imbalanced bankruptcy prediction models. International Journal on Electrical Engineering & Informatics 10 (3):433–46. doi:10.15676/ijeei.2018.10.3.2.

 Google Scholar

Zelenkov, Y., and N. Volodarskiy. 2021. Bankruptcy prediction on the base of the unbalanced data using multi-objective selection of classifiers. Expert Systems with Applications 185:115559. doi:10.1016/j.eswa.2021.115559.

 Web of Science ®Google Scholar

Zhao, S., K. Xu, Z. Wang, C. Liang, W. Lu, and B. Chen. 2022. Financial distress prediction by combining sentiment tone features. Economic modelling 106:105709. doi:10.1016/j.econmod.2021.105709.

 Web of Science ®Google Scholar

Le, T., B. Vo, H. Fujita, N.-T. Nguyen, and S. W. Baik. 2019. A fast and accurate approach for bankruptcy forecasting using squared logistics loss with GPU-based extreme gradient boosting. Information Sciences 494:294–310. doi:10.1016/j.ins.2019.04.060.

 Web of Science ®Google Scholar

Sun, J., H. Li, H. Fujita, B. Fu, and W. Ai. 2020. Class-imbalanced dynamic financial distress prediction based on Adaboost-SVM ensemble combined with SMOTE and time weighting. Information Fusion 54:128–44. doi:10.1016/j.inffus.2019.07.006.

 Web of Science ®Google Scholar

Du, X., W. Li, S. Ruan, and L. Li. 2020. CUS-heterogeneous ensemble-based financial distress prediction for imbalanced dataset with ensemble feature selection. Applied Soft Computing 97:106758. doi:10.1016/j.asoc.2020.106758.

 Web of Science ®Google Scholar

Huang, Y.-P., and M.-F. Yen. 2019. A new perspective of performance comparison among machine learning algorithms for financial distress prediction. Applied Soft Computing 83:105663. doi:10.1016/j.asoc.2019.105663.

 Web of Science ®Google Scholar

Son, H., C. Hyun, D. Phan, and H. Hwang. 2019. Data analytic approach for bankruptcy prediction. Expert Systems with Applications 138:112816. doi:10.1016/j.eswa.2019.07.033.

 Web of Science ®Google Scholar

Volkov, A., D. F. Benoit, and D. Van den Poel. 2017. Incorporating sequential information in bankruptcy prediction with predictors based on Markov for discrimination. Decision Support Systems 98:59–68. doi:10.1016/j.dss.2017.04.008.

 Web of Science ®Google Scholar

Jones, S. 2017. Corporate bankruptcy prediction: A high dimensional analysis. Review of Accounting Studies 22 (3):1366–422. doi:10.1007/s11142-017-9407-1.

 Web of Science ®Google Scholar

Kumar, G., and S. Roy. (2016). Development of hybrid boosting technique for bankruptcy prediction. In 2016 International Conference on Information Technology (ICIT), Bhubaneswar, India, pp. 248–53.

 Google Scholar

Pavlicko, M., M. Durica, and J. Mazanec. 2021. Ensemble model of the financial distress prediction in visegrad group countries. Mathematics 9 (16):1886. doi:10.3390/math9161886.

 Web of Science ®Google Scholar

Kanojia, S., and S. Gupta. 2022. Bankruptcy in Indian context: Perspectives from corporate governance. Journal of Management and Governance. doi:10.1007/s10997-022-09630-z.

 Web of Science ®Google Scholar

Putri, H. R., and A. Dhini. (2019). Prediction of financial distress: analyzing the industry performance in stock exchange market using data mining. In 2019 16th International Conference on Service Systems and Service Management (ICSSSM), Shenzhen, China, pp. 1–5.

 Google Scholar

Behr, A., and J. Weinblat. 2017. Default prediction using balance-sheet data: A comparison of models. The Journal of Risk Finance 18 (5):523–40. doi:10.1108/JRF-01-2017-0003.

 Web of Science ®Google Scholar

Jiang, C., X. Lyu, Y. Yuan, Z. Wang, and Y. Ding. 2021. Mining semantic features in current reports for financial distress prediction: Empirical evidence from unlisted public firms in China. International Journal of Forecasting 38 (3):1086–99. doi:10.1016/j.ijforecast.2021.06.011.

 Web of Science ®Google Scholar

Nyitrai, T. 2019. Dynamization of bankruptcy models via indicator variables. Benchmarking: An International Journal 26 (1):317–32. doi:10.1108/BIJ-03-2017-0052.

 Web of Science ®Google Scholar

Andreano, M. S., R. Benedetti, A. Mazzitelli, F. Piersimoni, and D. Di Fatta. 2018. How interconnected SMEs in business cluster survive the economic crisis. Kybernetes 50 (7). doi:10.1108/K-06-2018-0282.

 Google Scholar

Al-Milli, N., A. Hudaib, and N. Obeid. 2021. Population diversity control of genetic algorithm using a novel injection method for bankruptcy prediction problem. Mathematics 9 (8):823. doi:10.3390/math9080823.

 Web of Science ®Google Scholar

Huang, C., Q. Yang, M. Du, and D. Yang. 2017. Financial distress prediction using SVM ensemble based on earnings manipulation and fuzzy integral. Intelligent Data Analysis 21 (3):617–36. doi:10.3233/IDA-160034.

 Web of Science ®Google Scholar

Kim, S. Y., and A. Upneja. 2021. Majority voting ensemble with a decision trees for business failure prediction during economic downturns. Journal of Innovation & Knowledge 6 (2):112–23. doi:10.1016/j.jik.2021.01.001.

 Web of Science ®Google Scholar

UlagaPriya, K., and S. Pushpa. (2021). A comprehensive study on ensemble-based imbalanced data classification methods for bankruptcy data. In 2021 6th International Conference on Inventive Computation Technologies (ICICT), Coimbatore, India, pp. 800–04. IEEE.

 Google Scholar

Altman, E. I. 1968. Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of Finance 23 (4):589–609. doi:10.1111/j.1540-6261.1968.tb00843.x.

 Web of Science ®Google Scholar

Ptak-Chmielewska, A. 2021. Bankruptcy prediction of small- and medium-sized enterprises in Poland based on the LDA and SVM methods. Statistics in Transition New Series 22 (1):179–95. doi:10.21307/stattrans-2021-010.

 Google Scholar

Rasmussen, C. E., and C. K. I. Williams. 2006. Gaussian processes for machine learning, Cambridge, Massachusetts Institute of Technology. London, England: The MIT Press.

 Google Scholar

Antunes, F., B. Ribeiro, and F. Pereira. 2017. Probabilistic modeling and visualization for bankruptcy prediction. Applied Soft Computing 60:831–43. doi:10.1016/j.asoc.2017.06.043.

 Web of Science ®Google Scholar

Yang, Y., and C. Yang. (2020). Research on the application of GA improved neural network in the prediction of financial crisis. In 2020 12th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA), Phuket, Thailand, pp. 625–29.

 Google Scholar

Alam, N., J. Gao, and S. Jones. 2021. Corporate failure prediction: An evaluation of deep learning vs discrete hazard models. Journal of International Financial Markets, Institutions and Money 75:101455. doi:10.1016/j.intfin.2021.101455.

 Web of Science ®Google Scholar

Escribano-Navas, M., and G. Gemar. 2021. Gender and bankruptcy: a hotel survival econometric analysis. Sustainability 13 (12):6782. doi:10.3390/su13126782.

 Web of Science ®Google Scholar

Tobback, E., J. Moeyersoms, M. Stankova, and D. Martens. 2017. Bankruptcy prediction for SMEs using relational data. Decision Support Systems 102:69–81. doi:10.1016/j.dss.2017.07.004.

 Web of Science ®Google Scholar

Ouenniche, J., B. Pérez-Gladish, and K. Bouslah. 2018. An out-of-sample framework for TOPSIS-based classifiers with application in bankruptcy prediction. Technological Forecasting and Social Change 131:111–16. doi:10.1016/j.techfore.2017.05.034.

 Web of Science ®Google Scholar

Rahayu, D. S., and H. Suhartanto. (2020). Financial distress prediction in indonesia stock exchange’s listed company using case based reasoning concept. In 2020 IEEE 7th International Conference on Industrial Engineering and Applications (ICIEA), Bangkok, Thailand, pp. 1009–13.

 Google Scholar

Dumitrescu, A., M. El Hefnawy, and M. Zakriya. 2020. Golden geese or black sheep: Are stakeholders the saviors or saboteurs of financial distress? Finance Research Letters 37:101371. doi:10.1016/j.frl.2019.101371.

 Web of Science ®Google Scholar

Kovermann, J. H. 2018. Tax avoidance, tax risk and the cost of debt in a bank-dominated economy. Managerial Auditing Journal 33 (8/9):683–99. doi:10.1108/MAJ-12-2017-1734.

 Web of Science ®Google Scholar

Aalbers, H., J. Adriaanse, G.-J. Boon, J.-P.V.D Rest, R. Vriesendorp, and F. V. Wersch. 2019. Does pre-packed bankruptcy create value? An empirical study of postbankruptcy employment retention in the Netherlands. International Insolvency Review 28 (3):320–3571. doi:10.1002/iir.1353.

 Web of Science ®Google Scholar

Boubaker, S., A. Cellier, R. Manita, and A. Saeed. 2020. Does corporate social responsibility reduce financial distress risk? Economic modelling 91:835–51. doi:10.1016/j.econmod.2020.05.012.

 Web of Science ®Google Scholar

Liu, B., T. Ju, M. Bai, and C.-F.J Yu. 2021. Imitative innovation and financial distress risk: The moderating role of executive foreign experience. International Review of Economics & Finance 71:526–48. doi:10.1016/j.iref.2020.09.021.

 Web of Science ®Google Scholar

Mathew, S., S. Ibrahim, and S. Archbold. 2016. Boards attributes that increase firm risk – evidence from the UK. Corporate Governance 16 (2):233–58. doi:10.1108/CG-09-2015-0122.

 Web of Science ®Google Scholar

Hernandez Tinoco, M., P. Holmes, and N. Wilson. 2018. Polytomous response financial distress models: The role of accounting, market and macroeconomic variables. International Review of Financial Analysis 59:276–89. doi:10.1016/j.irfa.2018.03.017.

 Web of Science ®Google Scholar

Ali, S., R. Ur Rehman, W. Yuan, M. I. Ahmad, and R. Ali. 2021. Does foreign institutional ownership mediate the nexus between board diversity and the risk of financial distress? A case of an emerging economy of China. Eurasian Business Review 12 (3):553–81. doi:10.1007/s40821-021-00191-z.

 Web of Science ®Google Scholar

Succurro, M., G. Arcuri, and G. D. Costanzo. 2019. A combined approach based on robust PCA to improve bankruptcy forecasting. Review of Accounting and Finance 18 (2):296–320. doi:10.1108/RAF-04-2018-0077.

 Web of Science ®Google Scholar

Udin, S., M. A. Khan, and A. Y. Javid. 2017. The effects of ownership structure on likelihood of financial distress: An empirical evidence. Corporate Governance: The International Journal of Business in Society 17 (4):589–612. doi:10.1108/CG-03-2016-0067.

 Web of Science ®Google Scholar

Ben Jabeur, S. 2017. Bankruptcy prediction using partial least squares logistic regression. Journal of Retailing and Consumer Services 36:197–202. doi:10.1016/j.jretconser.2017.02.005.

 Web of Science ®Google Scholar

Balasubramanian, S. A., G.S. Radhakrishna, P. Sridevi, and T. Natarajan. 2019. Modeling corporate financial distress using financial and non-financial variables: The case of Indian listed companies. International Journal of Law and Management 61 (3/4):457–84. doi:10.1108/IJLMA-04-2018-0078.

 Web of Science ®Google Scholar

Cooper, E., and H. Uzun. 2019. Corporate social responsibility and bankruptcy. Studies in Economics and Finance 36 (2):130–53. doi:10.1108/SEF-01-2018-0013.

 Web of Science ®Google Scholar

Oz, I. O., and C. Simga-Mugan. 2018. Bankruptcy prediction models’ generalizability: Evidence from emerging market economies. Advances in Accounting 41:114–25. doi:10.1016/j.adiac.2018.02.002.

 Web of Science ®Google Scholar

Ahmad, A. H. 2019. What factors discriminate reorganized and delisted distressed firms: evidence from Malaysia. Finance Research Letters 29:50–56. doi:10.1016/j.frl.2019.03.010.

 Web of Science ®Google Scholar

Ashraf, S., E. G. S. Félix, and Z. Serrasqueiro. 2021. Does board committee independence affect financial distress likelihood? A comparison of China with the UK. Asia Pacific Journal of Management 39 (2):723–61. doi:10.1007/s10490-020-09747-5.

 Web of Science ®Google Scholar

Awwad, B., and B. Razia. 2021. Adapting Altman’s model to predict the performance of the Palestinian industrial sector. Journal of Business and Socio-Economic Development 1 (2):149–64. doi:10.1108/JBSED-05-2021-0063.

 Google Scholar

Bravo-Urquiza, F., and E. Moreno-Ureba. 2021. Does compliance with corporate governance codes help to mitigate financial distress? Research in International Business and Finance 55:101344. doi:10.1016/j.ribaf.2020.101344.

 Web of Science ®Google Scholar

Cenciarelli, V. G., G. Greco, and M. Allegrini. 2018. Does intellectual capital help predict bankruptcy? Journal of Intellectual Capital 19 (2):321–37. doi:10.1108/JIC-03-2017-0047.

 Web of Science ®Google Scholar

Chiou, K.-C., M.-M. Lo, and G.-W. Wu. (2017). The minimizing prediction error on corporate financial distress forecasting model: An application of dynamic distress threshold value. In 2017 IEEE 8th International Conference on Awareness Science and Technology (iCAST), Taichung, Taiwan, pp. 514–17.

 Google Scholar

Crosato, L., J. Domenech, and C. Liberati. 2021. Predicting SME’s default: Are their websites informative? Economics Letters 204:109888. doi:10.1016/j.econlet.2021.109888.

 Web of Science ®Google Scholar

Darrat, A. F., S. Gray, J. C. Park, and Y. Wu. 2016. Corporate governance and bankruptcy risk. Journal of Accounting, Auditing & Finance 31 (2):163–202. doi:10.1177/0148558X14560898.

 Web of Science ®Google Scholar

Doğan, S., D. Koçak, and M. Atan. 2022. Financial distress prediction using support vector machines and logistic regression. In Advances in econometrics, operational research, data science and actuarial studies: techniques and theories, contributions to economics, ed. M. K. Terzioğlu, 429–52. Cham: Springer International Publishing.

 Google Scholar

Freitas Cardoso, G., F. M. Peixoto, and F. Barboza. 2019. Board structure and financial distress in Brazilian firms. International Journal of Managerial Finance 15 (5):813–28. doi:10.1108/IJMF-12-2017-0283.

 Web of Science ®Google Scholar

García, C. J., and B. Herrero. 2021. Female directors, capital structure, and financial distress. Journal of Business Research 136:592–601. doi:10.1016/j.jbusres.2021.07.061.

 Web of Science ®Google Scholar

Jindal, N. 2020. The impact of advertising and R&D on bankruptcy survival: a double-edged sword. Journal of Marketing 84 (5):22–40. doi:10.1177/0022242920936205.

 Web of Science ®Google Scholar

Kamalirezaei, H., A. A. A. Rostamy, A. Saeedi, and M. K. V. Zaghard. 2020. Corporate social responsibility and bankruptcy probability: Exploring the role of market competition, intellectual capital, and equity cost. Journal of Corporate Accounting & Finance 31 (1):53–63. doi:10.1002/jcaf.22417.

 Web of Science ®Google Scholar

Kim, H., H. Cho, and D. Ryu. 2021. Corporate bankruptcy prediction using machine learning methodologies with a focus on sequential data. Computational Economics 59 (3):1231–49. doi:10.1007/s10614-021-10126-5.

 Web of Science ®Google Scholar

Li, L., and R. Faff. 2019. Predicting corporate bankruptcy: What matters? International Review of Economics & Finance 62:1–19. doi:10.1016/j.iref.2019.02.016.

 Web of Science ®Google Scholar

Lin, K. C., and X. Dong. 2018. Corporate social responsibility engagement of financially distressed firms and their bankruptcy likelihood. Advances in Accounting 43:32–45. doi:10.1016/j.adiac.2018.08.001.

 Web of Science ®Google Scholar

Mai, F., S. Tian, C. Lee, and L. Ma. 2019. Deep learning models for bankruptcy prediction using textual disclosures. European Journal of Operational Research 274 (2):743–58. doi:10.1016/j.ejor.2018.10.024.

 Web of Science ®Google Scholar

Olsen, B. C., and C. Tamm. 2017. Corporate governance changes around bankruptcy. Managerial Finance 43 (10):1152–69. doi:10.1108/MF-09-2015-0257.

 Web of Science ®Google Scholar

Park, S., H. Kim, J. Kwon, and T. Kim. 2021. Empirics of Korean shipping companies’ default predictions. Risks 9 (9):159. doi:10.3390/risks9090159.

 Web of Science ®Google Scholar

Pham Vo Ninh, B., T. Do Thanh, and D. Vo Hong. 2018. Financial distress and bankruptcy prediction: An appropriate model for listed firms in Vietnam. Economic Systems 42 (4):616–24. doi:10.1016/j.ecosys.2018.05.002.

 Web of Science ®Google Scholar

Ravula, S. 2021. Bankruptcy prediction using disclosure text features. arXiv:2101.00719 [cs, q-fin]. 10.1111/ijcp.14016

 Google Scholar

Regenburg, K., and M. N. B. Seitz. 2021. Criminals, bankruptcy, and cost of debt. Review of Accounting Studies 26 (3):1004–45. doi:10.1007/s11142-021-09608-6.

 Web of Science ®Google Scholar

Ren, T., T. Lu, and Y. Yang. (2021). Improved data mining method for class-imbalanced financial distress prediction. In 2021 7th International Conference on Computing and Artificial Intelligence, Tianjin China, pp. 308–13.

 Google Scholar

Richardson, G., R. Lanis, and G. Taylor. 2015. Financial distress, outside directors and corporate tax aggressiveness spanning the global financial crisis: An empirical analysis. Journal of Banking & Finance 52:112–29. doi:10.1016/j.jbankfin.2014.11.013.

 Web of Science ®Google Scholar

Sayari, N., and C. S. Mugan. 2017. Industry specific financial distress modeling. BRQ Business Research Quarterly 20 (1):45–62. doi:10.1016/j.brq.2016.03.003.

 Web of Science ®Google Scholar

Shahwan, T. M., and A. M. Habib. 2020. Does the efficiency of corporate governance and intellectual capital affect a firm’s financial distress? Evidence from Egypt. Journal of Intellectual Capital 21 (3):403–30. doi:10.1108/JIC-06-2019-0143.

 Web of Science ®Google Scholar

Stef, N., and E. Zenou. 2021. Management-to-staff ratio and a firm’s exit. Journal of Business Research 125:252–60. doi:10.1016/j.jbusres.2020.12.027.

 Web of Science ®Google Scholar

Süsi, V., and O. Lukason. 2019. Corporate governance and failure risk: Evidence from Estonian SME population. Management Research Review 42 (6):703–20. doi:10.1108/MRR-03-2018-0105.

 Web of Science ®Google Scholar

Yazdanfar, D., and P. Öhman. 2020. Financial distress determinants among SMEs: Empirical evidence from Sweden. Journal of Economic Studies 47 (3):547–60. doi:10.1108/JES-01-2019-0030.

 Web of Science ®Google Scholar

Angenent, M. N., A. P. Barata, and F. W. Takes. (2020). Large-scale machine learning for business sector prediction. In Proceedings of the 35th Annual ACM Symposium on Applied Computing, SAC ’20, New York, NY, USA, pp. 1143–46.

 Google Scholar

Chang, H. (2019). The application of machine learning models in company bankruptcy prediction. In Proceedings of the 2019 3rd International Conference on Software and e-Business, ICSEB 2019, New York, NY, USA, pp. 199–203.

 Google Scholar

Joshi, S., R. Ramesh, and S. Tahsildar. (2018). A bankruptcy prediction model using random forest. In 2018 Second International Conference on Intelligent Computing and Control Systems (ICICCS), Madurai, India, pp. 1–6.

 Google Scholar

Lu, Y., J. Zhu, N. Zhang, and Q. Shao. (2014). A hybrid switching PSO algorithm and support vector machines for bankruptcy prediction. In 2014 International Conference on Mechatronics and Control (ICMC), Jinzhou, China, pp. 1329–33.

 Google Scholar

Sun, J., H. Fujita, Y. Zheng, and W. Ai. 2021. Multi-class financial distress prediction based on support vector machines integrated with the decomposition and fusion methods. Information Sciences 559:153–70. doi:10.1016/j.ins.2021.01.059.

 Web of Science ®Google Scholar

Tsai, M.-F., and C.-J. Wang. 2017. On the risk prediction and analysis of soft information in finance reports. European Journal of Operational Research 257 (1):243–50. doi:10.1016/j.ejor.2016.06.069.

 Web of Science ®Google Scholar

Zoričák, M., P. Gnip, P. Drotár, and V. Gazda. 2020. Bankruptcy prediction for small- and medium-sized companies using severely imbalanced datasets. Economic modelling 84:165–76. doi:10.1016/j.econmod.2019.04.003.

 Web of Science ®Google Scholar

Ouenniche, J., K. Bouslah, B. Perez-Gladish, and B. Xu. 2021. A new VIKOR-based in-sample-out-of-sample classifier with application in bankruptcy prediction. Annals of Operations Research 296 (1–2):495–512. doi:10.1007/s10479-019-03223-0.

 Web of Science ®Google Scholar

Ahmadi, Z., P. Martens, C. Koch, T. Gottron, and S. Kramer. (2018). Towards bankruptcy prediction: deep sentiment mining to detect financial distress from business management reports. In 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA), Turin, Italy, pp. 293–302.

 Google Scholar

Abdi, A., S. M. Shamsuddin, S. Hasan, and J. Piran. 2019. Deep learning-based sentiment classification of evaluative text based on Multi-feature fusion. Information Processing & Management 56 (4):1245–59. doi:10.1016/j.ipm.2019.02.018.

 Web of Science ®Google Scholar

Xu, W., H. Fu, and Y. Pan. 2019. A novel soft ensemble model for financial distress prediction with different sample sizes. Mathematical Problems in Engineering 2019:1–12. doi: 10.1155/2019/3085247

 Web of Science ®Google Scholar

Wang, M., H. Chen, H. Li, Z. Cai, X. Zhao, C. Tong, J. Li, and X. Xu. 2017. Grey wolf optimization evolving kernel extreme learning machine: Application to bankruptcy prediction. Engineering Applications of Artificial Intelligence 63:54–68. doi:10.1016/j.engappai.2017.05.003.

 Web of Science ®Google Scholar

Zhang, Y., R. Liu, A. A. Heidari, X. Wang, Y. Chen, M. Wang, and H. Chen. 2021. Towards augmented kernel extreme learning models for bankruptcy prediction: Algorithmic behavior and comprehensive analysis. Neurocomputing 430:185–212. doi:10.1016/j.neucom.2020.10.038.

 Web of Science ®Google Scholar

Chou, C.-H., S.-C. Hsieh, and C.-J. Qiu. 2017. Hybrid genetic algorithm and fuzzy clustering for bankruptcy prediction. Applied Soft Computing 56:298–316. doi:10.1016/j.asoc.2017.03.014.

 Web of Science ®Google Scholar