|
| Título : |
Perfilado de conductores mediante técnicas de ML |
| Otro título : |
Driver profiling through Machine Learning techniques |
| Tipo de documento: |
documento electrónico |
| Autores: |
José López Galdón, Autor ; Hugo César Octavio del Sueldo, Autor ; Juan Manuel López Zafra, Director de tesi |
| Fecha de publicación: |
2021 |
| Número de páginas: |
28 p. |
| Nota general: |
Máster en Data Science para Finanzas |
| Idioma : |
Inglés (eng) |
| Materias: |
Inteligencia artificial
Seguros de accidentes
Tratamiento automático de datos
|
| Palabras clave: |
Insurance; data science; supervised and unsupervised algorithms; machine learning; clustering; decision tree. |
| Clasificación: |
004.8 Inteligencia artificial. Razonamiento y aprendizaje automatizados. Sistemas inteligentes |
| Resumen: |
Traffic accidents represent a high cost for insurance companies as well as for society, in economic and social terms, because in all cases the costs include medical and rehabilitation expenses, legal and emergency services, property damage and production losses. Thanks to the use of telematics and data science we may be able to find patterns of behavior that explain the claims. During this research we will work with a database of more than 95,000 drivers that includes information collected over 6 years; for this, we have performed an important work of cleaning and engineering of variables, for finally clustering the drivers through a PAM, being the most representative variables the intensity of use of the vehicle and the driving experience. In addition, we have made a prediction based on whether or not they have suffered a crash using a decision tree, obtaining a 72.25% accuracy rate. |
| Link: |
https://biblioteca.cunef.edu/gestion/catalogo/index.php?lvl=notice_display&id=48904 |
Perfilado de conductores mediante técnicas de ML = Driver profiling through Machine Learning techniques [documento electrónico] / José López Galdón, Autor ; Hugo César Octavio del Sueldo, Autor ; Juan Manuel López Zafra, Director de tesi . - 2021 . - 28 p. Máster en Data Science para Finanzas Idioma : Inglés ( eng)
| Materias: |
Inteligencia artificial
Seguros de accidentes
Tratamiento automático de datos
|
| Palabras clave: |
Insurance; data science; supervised and unsupervised algorithms; machine learning; clustering; decision tree. |
| Clasificación: |
004.8 Inteligencia artificial. Razonamiento y aprendizaje automatizados. Sistemas inteligentes |
| Resumen: |
Traffic accidents represent a high cost for insurance companies as well as for society, in economic and social terms, because in all cases the costs include medical and rehabilitation expenses, legal and emergency services, property damage and production losses. Thanks to the use of telematics and data science we may be able to find patterns of behavior that explain the claims. During this research we will work with a database of more than 95,000 drivers that includes information collected over 6 years; for this, we have performed an important work of cleaning and engineering of variables, for finally clustering the drivers through a PAM, being the most representative variables the intensity of use of the vehicle and the driving experience. In addition, we have made a prediction based on whether or not they have suffered a crash using a decision tree, obtaining a 72.25% accuracy rate. |
| Link: |
https://biblioteca.cunef.edu/gestion/catalogo/index.php?lvl=notice_display&id=48904 |
|
Refinar búsquedaAlquimia / Juan Manuel López Zafra (2019)
