Version 1
: Received: 23 September 2024 / Approved: 23 September 2024 / Online: 24 September 2024 (00:18:52 CEST)
How to cite:
Chai, E.; Hadullo, K.; Tole, K.; Stephen, D. N. A Dual-Phase Framework for Enhanced Churn Prediction in Motor Insurance using Cave-Degree and Magnetic Force Perturbation Techniques. Preprints2024, 2024091820. https://doi.org/10.20944/preprints202409.1820.v1
Chai, E.; Hadullo, K.; Tole, K.; Stephen, D. N. A Dual-Phase Framework for Enhanced Churn Prediction in Motor Insurance using Cave-Degree and Magnetic Force Perturbation Techniques. Preprints 2024, 2024091820. https://doi.org/10.20944/preprints202409.1820.v1
Chai, E.; Hadullo, K.; Tole, K.; Stephen, D. N. A Dual-Phase Framework for Enhanced Churn Prediction in Motor Insurance using Cave-Degree and Magnetic Force Perturbation Techniques. Preprints2024, 2024091820. https://doi.org/10.20944/preprints202409.1820.v1
APA Style
Chai, E., Hadullo, K., Tole, K., & Stephen, D. N. (2024). A Dual-Phase Framework for Enhanced Churn Prediction in Motor Insurance using Cave-Degree and Magnetic Force Perturbation Techniques. Preprints. https://doi.org/10.20944/preprints202409.1820.v1
Chicago/Turabian Style
Chai, E., Kevin Tole and Dorca Nyamusi Stephen. 2024 "A Dual-Phase Framework for Enhanced Churn Prediction in Motor Insurance using Cave-Degree and Magnetic Force Perturbation Techniques" Preprints. https://doi.org/10.20944/preprints202409.1820.v1
Abstract
This study presents a novel predictive model to address the churn problem in the motor insurance sector using a dual-phase framework. The first phase employs a cave-degree perturbation technique for effective feature selection, while the second phase applies the Magnetic Force Perturbation Technique (MFPT) to optimize the search process and avoid local optima traps. Two metaheuristics are proposed: the Adaptive Random Forest-Assisted Large Neighborhood Feature Optimizer (ARALFO) and the Adaptive Random Forest Particle Swarm Optimizer (ARFPSO) to enhance churn prediction accuracy. The model was evaluated on two real-world motor insurance datasets, achieving a 95% accuracy rate and outperforming state-of-the-art algorithms. An ablation study confirmed the significant impact of the cave-degree and MFPT techniques in boosting predictive performance.
Keywords
Churn Prediction; Insurance Motor; Algorithm; Cave degree; Magnetic Force
Subject
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
