Credit Approval System Using Machine Learning: Challenges and Future Directions

Abstract

The applications of machine learning have now reached variety of industries, including banking and financial organisations. While credit approval is a key concern of the banking industry, machine learning is widely regarded as one of the most effective methods for credit approval. In fact, due to significant amount of research being conducted in this domain, enhanced new algorithms and/or approaches are continuously being proposed by the researchers. Therefore, to compare, contrast and synthesise the performance of these machine learning algorithms, this literature survey covered 52 articles since as far back as 2000. We have also recommended the application of fuzziness-based semi-supervised learning, which has never been previously utilised in the credit approval process, as per our survey findings.