Centro de Documentación y Conocimiento de Fundación Mapfre > Poisson mixture deep learning neural network models for the prediction of drivers' claims with excessive zero claims using telematics data

Kelly, Mary

Poisson mixture deep learning neural network models for the prediction of drivers' claims with excessive zero claims using telematics data / Farha Usman ... [et al.]

Sumario: This article develops Poisson Mixture Deep Learning Neural Network models to improve the prediction of automobile insurance claim frequency using telematics data. The study addresses the challenge of an excessive proportion of policyholders with zero claims through advanced machine learning techniques, driver risk classification, and predictive model optimization. The results demonstrate improvements in driver segmentation and in personalized insurance pricing based on driving behavior and usage

En: North American actuarial journal. - Schaumburg : Society of Actuaries, 1997- = ISSN 1092-0277. - 21/09/2026 Volume 30 Number 3 - 2026 , 22 p.

1. Reclamaciones . 2. Mercado de seguros . 3. Seguro de automóviles . 4. Siniestros . 5. Big data . 6. Inteligencia artificial . 7. Modelos predictivos . 8. Primas de seguros . 9. Estados Unidos . I. Chan, Jennifer S.K. . II. Dong, Alice X. D. . III. Makov, Udi E. . IV. Título.