Análisis de tweets y su influencia en los seguros de vida en el ámbito colombiano

In recent years, life insurance has increased its participation in the Colombian financial market. To strengthen this position, insurers have the challenge of reinventing their products adjusting to the demands of their clients and thus increasing sales. Thus, to fully encourage the development of life insurance and streamline its offer in the country, in this master's work a natural language processing model trained in the profiling task is designed, developed and evaluated. unstructured data from Twitter. With these data, a segmentation of users is generated by analyzing eating habits, physical activity and the sentiment of the publications, allowing to offer a preferential rate for the fragments identified with less risk that generates a stimulus in the sale and creates economic value for insurers. The model is validated by relating the results to the mortality rate at the departmental level, obtaining a moderate Pearson's correlation of 0.56 for the urban departments of Colombia.

• Autores: Ávila Rodríguez, Maria Paula
• Fecha de publicación: 2020
• Institución: Escuela Colombiana de Ingeniería Julio Garavito
• Repositorio: Repositorio Institucional ECI
• Idioma: spa
• OAI Identifier: oai:repositorio.escuelaing.edu.co:001/1237
• Acceso en línea: https://catalogo.escuelaing.edu.co/cgi-bin/koha/opac-detail.pl?biblionumber=22429; https://repositorio.escuelaing.edu.co/handle/001/1237
• Palabra clave: Tarificación; Seguros de vida; Minería de datos; Procesamiento de lenguaje natural-Colombia; Pricing; life insurance; data mining; natural language processing- Colombia
• Rights: openAccess
• License: Derechos Reservados - Escuela Colombiana de Ingeniería Julio Garavito