Predicting Credit Card Approval of Customers Through Customer Profiling using Machine Learning
Arokiaraj Christian St Hubert1, R. Vimalesh2, M. Ranjith3, S. Aravind Raj4
1Arokiaraj Christian St Hubert*, Computer Science and Engineering, Sri Manakula Vinayagar Engineering College, Puducherry, India.
2R. Vimalesh, Computer Science and Engineering, Sri Manakula Vinayagar Engineering College, Puducherry, India.
3M. Ranjith, Computer Science and Engineering, Sri Manakula Vinayagar Engineering College, Puducherry, India.
4S. Aravind Raj, Computer Science and Engineering, Sri Manakula Vinayagar Engineering College, Puducherry, India.
Manuscript received on March 05, 2020. | Revised Manuscript received on March 16, 2020. | Manuscript published on April 30, 2020. | PP: 552-557 | Volume-9 Issue-4, April 2020. | Retrieval Number: D7293049420/2020©BEIESP | DOI: 10.35940/ijeat.D7293.049420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In the banking sector, every banking infrastructure contains an enormous dataset for customers’ credit card approval which requires customer profiling. The customer profiling means collection of data related to what customers need. It depends on customers’ basic information like field of work, address proof, credit score, salary details, etc. This process mainly concentrates on predicting approval of credit cards to customers using machine learning. Machine Learning is the scientific study of algorithms and statistical models that computers use to perform specific tasks without any external instructions or interference. In the current trend this process is possible using many algorithms like “K-Mean, Improved K-Mean and Fuzzy C-Means”. This helps banks to have an high profitability to satisfy their customers. However, the currently prevailing system shows an accuracy percentage of about 98.08%. The proposed system aims at improvising the accuracy ratio while using only few algorithms.
Keywords: Credit Score, Machine Learning, Supervised Learning, Unsupervised Learning.