PENGGUNAAN MODEL ARTIFICIAL INTELLIGENCE DAN MACHINE LEARNING PADA PREDIKSI KEBANGKRUTAN – A SYSTEMATIC LITERATURE REVIEW

Abstract

This study examines the use of Artificial Intelligence (AI) and Machine Learning (ML) in corporate bankruptcy prediction through a Systematic Literature Review (SLR). The review analyzes 30 Scopus-indexed articles published between 2021 and 2025. The findings show that bankruptcy prediction has shifted from traditional statistical models to more adaptive and accurate AI/ML approaches, with dominant models including Random Forest, Gradient Boosting, LightGBM, SVM, ANN, and DNN. The significant predictors include not only financial ratios but also non-financial factors such as corporate governance and financial reporting quality. The study also identifies key challenges, including imbalanced data, overfitting, feature selection, and limited interpretability, which can be addressed through data balancing, feature selection, and explainable AI techniques. Overall, AI and ML have strong potential to improve bankruptcy prediction effectiveness when supported by high-quality data and appropriate model selection.