Forecasting Financial Distress With Machine Learning – A Review

• Autor(a) principal: Duarte, Denize Lemos
• Data de Publicação: 2020
• Outros Autores: Barboza, Flávio Luiz de Moraes
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Future Studies Research Journal: Trends and Strategies
• Texto Completo: https://www.revistafuture.org/FSRJ/article/view/533
• topic: Bankruptcy; Credit Risk; Artificial Intelligence; Machine Learning

Purpose – Evaluate the various academic researches with multiple views on credit risk and artificial intelligence (AI) and their evolution.Theoretical framework – The study is divided as follows: Section 1 introduces the article. Section 2 deals with credit risk and its relationship with computational models and techniques. Section 3 presents the methodology. Section 4 addresses a discussion of the results and challenges on the topic. Finally, section 5 presents the conclusions.Design/methodology/approach – A systematic review of the literature was carried out without defining the time period and using the Web of Science and Scopus database.Findings – The application of computational technology in the scope of credit risk analysis has drawn attention in a unique way. It was found that the demand for identification and introduction of new variables, classifiers and more assertive methods is constant. The effort to improve the interpretation of data and models is intense.Research, Practical & Social implications – It contributes to the verification of the theory, providing information in relation to the most used methods and techniques, it brings a wide analysis to deepen the knowledge of the factors and variables on the theme. It categorizes the lines of research and provides a summary of the literature, which serves as a reference, in addition to suggesting future research.Originality/value – Research in the area of Artificial Intelligence and Machine Learning is recent and requires attention and investigation, thus, this study contributes to the opening of new views in order to deepen the work on this topic.