Comparing the Performance of Corporate Bankruptcy Prediction Models Based on Imbalanced Financial Data

Forecasts of corporate defaults are used in various fields across the economy. Several recent studies attempt to forecast bankruptcy using machine learning techniques. We collected financial information on 13 variables 1020 companies listed KOSPI and KOSDAQ capture possibility bankruptcy. propose a data processing method for small-sample domestic data. investigate case random sampling non-bankrupt versus based approximate entropy optimized threshold AUC address imbalance between number bankrupt companies. compare performance measures prediction models small sample structured two ways full dataset. The experimental results this study contribute selection an appropriate model.