Hernández Gress, NeilCadena Rodríguez, Rodrigo2025-03-122023-05-31Cadena Rodríguez, R. (2023). Risk factor classification for drivers in Mexico using data science [Tesis maestría]. Instituto Tecnológico y de Estudios Superiores de Monterrey. Recuperado de: https://hdl.handle.net/11285/703316https://hdl.handle.net/11285/703316The aim of this dissertation is to find an optimal way to profile drivers in Mexico analysing different databases of car accidents and auto insurance claims inside this country and using gradient boosting algorithms. According to the National Public Health Institute, Mexico is in seventh-place globally and third place in Latin America in the most deaths caused by car accidents' ranking. Moreover, even when it is mandatory to have car insurance when having a car, only 30\% of people hires a car insurance. This is mainly because of the prices that insurance companies offer, and this happens because most of them are using old methods that do not consider all the crucial variables and treat all their customers as if everybody had the same risk for making a claim, even when companies in other countries are using some machine learning models that have been proved to be efficient and permitted a low-cost premium based on users profile.TextoengopenAccesshttp://creativecommons.org/licenses/by/4.0INGENIERÍA Y TECNOLOGÍA::CIENCIAS TECNOLÓGICAS::TECNOLOGÍA DE LOS ORDENADORES::SISTEMAS DE CONTROL MÉDICOTechnologyTesis de Maestría / master Thesishttps://orcid.org/0009-0006-5080-3305Gradient BoostingMachine LearningXGBoostCatBoostLight GBMSMOTESMOTEENRisk Profiling