A Recommendation System For Insurance Built With A Multivariate Hawkes Process Based On Customers' Life Events

Laurent Lesage (1, 2, 3, 4) , Madalina Deaconu (2, 3) , Antoine Lejay (2, 3) , Jorge Augusto Meira (5) , Geoffrey Nichil (4) , Radu State (5)

1 SnT - Interdisciplinary Centre for Security, Reliability and Trust [Luxembourg]
2 PASTA - Processus aléatoires spatio-temporels et leurs applications
3 IECL - Institut Élie Cartan de Lorraine
4 Foyer Assurances [Leudelange]
5 University of Luxembourg [Luxembourg]

We construct a recommendation system for insurance with a Multivariate Hawkes Process, in order to take into account the influence of life events (e.g. marriage, birth, change of job) on the insurance covering selection from customers. This Multivariate Hawkes Process includes several specific features aiming to compute relevant recommendations to customers from a Luxembourgish insurance company. Some of these features are intent to propose a personalized background intensity for each customer thanks to a Machine Learning model, to use triggering functions suited for insurance data or to overcome flaws in real-world data by adding a specific penalization term in the objective function. Our recommendation system has been backtested over a full year. Observations from model parameters and results from this back-test show that taking into account life events by a Multivariate Hawkes Process allows us to improve significantly the accuracy of recommendations.

Mots clés: Insurance; Multivariate Hawkes Process; Recommendation System; Life Events

Domaines: Intelligence artificielle [cs.AI]

Laurent Lesage, Madalina Deaconu, Antoine Lejay, Jorge Augusto Meira, Geoffrey Nichil, et al.. A Recommendation System For Insurance Built With A Multivariate Hawkes Process Based On Customers' Life Events. 2021. ⟨hal-03483812⟩