Enhancing Credit Card Application Approval through Data Scaling in Machine Learning Algorithms

In today's fast-paced world, financial transactions have undergone a remarkable transformation, and at the forefront of this revolution stand the credit card. Over the past few years, the usage of credit cards has become ingrained in our day-to-day lives. Many financial institutions, such as banks, issue credit cards in response to applications. However, these applications need to undergo verification before approval to prevent fraudulence and potential losses for the institutions. Automation of the credit card application approval process is being increasingly adopted by many financial institutions as it saves time and mitigates the involvement of human bias. Nevertheless, challenges related to the dataset, such as missing values, inconsistency and numerical noise persists. If not adequately addressed, these issues can adversely impact the performance of Machine Learning (ML) algorithms. Data preprocessing plays a crucial role in enhancing the model's performance and involves essential steps like data conversion, feature scaling, feature selection and reduction. These steps significantly contribute to increasing the prediction accuracy. The proposed work aims to compare various traditional ML algorithms concerning different scaling methods. The results demonstrate that the proposed scaling methods (MinMaxScaler, MaxAbsScaler, StandardScaler, RobustScaler) outperform existing approaches, showing promise for the automated credit card application approval process.