Bankruptcy Prediction Using Deep Learning Approach Based on Borderline SMOTE

Abstract

Imbalanced classification on bankruptcy prediction is considered as one of the most important topics in financial institutions. In this context, various statistical and artificial intelligence methods have been proposed. Recently, deep learning algorithms are experiencing a resurgence of interest, and are widely used to build a prediction and classification models. To this end, we propose a novel deep learning-based approach called BSM-SAES. This approach combines Borderline Synthetic Minority oversampling technique (BSM) and Stacked AutoEncoder (SAE) based on the Softmax classifier. The aim is to develop an accurate and reliable bankruptcy prediction model which includes the features extraction process. To assess the classification performance of our proposed model, k-nearest neighbor, decision tree, support vector machine, and artificial neural network, C5.0 that are machine learning methods, are applied. We evaluate our proposed approach on the Polish imbalanced datasets. The obtained results confirm the efficiency of our proposed model compared to other machine learning models regarding predicting and classifying the financial status of a firm.

Prediction of the Bankruptcy of Slovak Companies Using Neural Networks with SMOTE

“…Per the recommendation of Zhou ( 2013) we have selected oversampling as a tool for the reduction of bias, variance, and noise of the bankruptcy model. Following the conclusions of Le et al (2018), Hardle et al (2019), Shrivastava, Jeyanthi, and Singh, S. (2020), Smiti et al (2020), andFaris et al (2020), we used the SMOTE (synthetic minority oversampling technique) as the algorithm for oversampling the bankruptcy related entries.…”

Empirical Evaluation of Noise Influence on Supervised Machine Learning Algorithms Using Intrusion Detection Datasets

“…During the tree construction, the training dataset is recursively divided into several subsets. e best split is often based on children impurity such as the entropy which is defined as follows [16,17]:…”

“…is paper uses multilayer perceptron (MLP) to implement ANN within the experiments. MLP is an ANN, usually trained using backpropagation algorithm [17,20]. e output of the hidden neuron i can be calculated as follows:…”

“…., n; W ij denotes the weight connecting input neuron j to hidden neuron i, b i is the bias for neuron i, and f 1 is the used transfer function. e output of the output neuron i can be calculated as follows [17]: where n k is the number of hidden neurons and V ij denotes the weight connecting hidden neuron i to the output neuron j. e goal of ANN is to minimize the total sum of squared error between the predicted output and the true output (ground truth) [16]:…”

Company bankruptcy prediction framework based on the most influential features using XGBoost and stacking ensemble learning

“…Borderline synthetic minority over-sampling technique (BSM) and stacked auto-encoder (SAE) based on the Soft-max classifier are proposed to solve the unbalanced classification of company bankruptcy prediction problems. This combination approach is considered more efficient than the combination of BSM with machine learning techniques and machine learning techniques without over-sampling [44].…”