Analysing the Evolution of Machine Learning Techniques in Forecasting Financial Distress and Their Implications for Sustainability: A Comprehensive Review and Content Analysis

Abstract

The prediction of such conditions as financial distress or corporate failure and bankruptcy has drawn much attention recently because it gives an early signal of economic problems that corporations are likely to encounter. This field of work is essential to companies, creditors, and investors since it provides information on potential risks and, therefore, potential requirements for funding solutions. Machine learning (ML) models using large databases and better computational techniques introduced a relatively new area of financial distress prediction. To fill this gap and because the ML-based financial distress prediction literature is fragmented, this study performs a systematic literature review to analyze the existing literature. Specifically, by conducting a content analysis of the credit risk study, this paper establishes themes that show how data preprocessing enhances predictive performance. It further discusses top ML models, evaluating their applicability and discussing their advantages and disadvantages. It discusses the impact of the model on sustainability. The paper also deeply analyses the relationship between financial distress prediction and sustainability to provide worthwhile advice to researchers, practical workers, and investors.