Title: Assessing credit card eligibility using the mathematical models for optimal detection: internet protocol technology

Authors: Jing Ge

Addresses: Chongqing Vocational and Technical University of Mechatronics, Chongqing, China

Abstract: Eligibility for a credit card is determined by several factors, including credit score, income, employment status, existing debt and credit history. This research investigates the use of advanced computational techniques to predict credit card eligibility. Specifically, the study employs Decision Tree Classifier (DTC) and Random Forest Classifier (RFC) models, enhanced by the Fitness Dependent Optimiser (FDO) and Chita Optimisation Algorithm (COA), to improve prediction accuracy. The hybrid models were tested and compared based on their ability to predict acceptance or rejection of credit card applications. Results indicate that the RFCO model excels under rejection conditions, achieving the highest accuracy and a recall value of 0.9582. The RFFD model follows closely behind the RFCO model, achieving a recall value of 0.9504. Additionally, the DTFD model outperforms the DTCO model, with recall values of 0.9426 and 0.9399, respectively. The standalone DTC and RFC models achieved recall values of 0.9347 and 0.9321.

Keywords: credit card; machine learning; dependent optimiser; Chita optimisation algorithm; decision tree classifier; random forest classifier.

DOI: 10.1504/IJIPT.2024.145125

International Journal of Internet Protocol Technology, 2024 Vol.17 No.2, pp.89 - 100

Received: 10 Dec 2024
Accepted: 26 Jan 2025
Published online: 19 Mar 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article