A Machine Learning-Based Approach to Improve Fraud Detection using ISSO and PKRR

Fraud detection has become increasingly crucial in various industries, particularly within financial sectors. This study introduces a novel machine learning framework that integrates ISSO with PKRR to enhance fraud detection capabilities. By combining ISSO’s optimization strengths with PKRR’s robust regression and partitioning capabilities, the proposed model significantly boosts detection rates and minimizes false positives. This integrated approach addresses existing limitations, offering a scalable, adaptive, and effective solution for safeguarding financial systems against evolving fraud threats. Preliminary results indicate that this framework outperforms traditional methods, highlighting its potential for widespread application in the financial industry.