Systematic Literature Review: Bankruptcy Prediction Menggunakan Teknik Machine Learning dan Deep Learning

Abstract

Bankruptcy prediction has long been an interesting topic to be discussed and developed by many researchers. Initially, the development of bankruptcy prediction models was carried out through statistical methods by examining corporate's financial ratios. Along with the development of technology, the use of machine learning and deep learning techniques for developing bankruptcy prediction models has begun to be widely used, for example, SVM, ANN, hybrid genetic algorithm, fuzzy-SVM, convolutional neural network, textual disclosure, and so on. Although machine learning and deep learning models have various advantages over statistical methods, they still have many drawbacks. In this paper, we will review machine learning and deep learning models used in bankruptcy prediction using the Systematic Literature Review method. Based on our research, bankruptcy prediction models developed using machine learning or deep learning techniques can outperform the performance of models using classical statistical methods.