Systematic Literature Review of Machine Learning Models used in the Analysis of Bank Data for Better Decision Making

Machine Learning (ML) Classification algorithms used in the Banking data set are reviewed in this paper. The algorithms performance and results may differ. The evaluation metrics used in the survey papers are measured and summarize each machine learning algorithm's performance. Therefore, in order to choose the best approach, a comparative study is needed to determine the most appropriate approach for a given domain. As we know, nowadays the banking business is based on the customer, so knowing profitable customer is essential for the better decision making in the banking business. The real-world of Bank is changed to customer-centricfromtheproduct- orientedones.In this paper, a study is presented between various classification methods, namely the linear machine learning algorithms and ensemble algorithms. The reviewed articles include various data set of different records and attributes consisting of demographic, behavioral and credit information of Bank from the primary sources and secondary sources such as Kaggle, UCI repositories etc. The paper concluded with the evaluation and comparative study of algorithms based on their classification reports. In the classification reports, the metrics of the algorithms has been computed. And also examined the confusion matrix of each machine learning algorithms.