Modelado basado en datos para la clasificación semiautomática de correspondencia electrónica: caso de estudio para la Administración Pública Colombiana
E-mail communication is still the most prevalent form of Customer support process in Organizations. Therefore, organizations have had to implementing processes focused on email categorization in accordance with email contents, in order to provide an efficient response to customer requests. A common...
- Autores:
-
Vargas Antolínez, Edwin Alberto
- Tipo de recurso:
- Fecha de publicación:
- 2018
- Institución:
- Escuela Colombiana de Ingeniería Julio Garavito
- Repositorio:
- Repositorio Institucional ECI
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.escuelaing.edu.co:001/829
- Acceso en línea:
- https://catalogo.escuelaing.edu.co/cgi-bin/koha/opac-detail.pl?biblionumber=21550
https://repositorio.escuelaing.edu.co/handle/001/829
- Palabra clave:
- Clasificación de correspondencia
Minería de datos
Aprendizaje de Máquina
Algoritmos de clasificación
Email classification
Data mining
Machine Learning
Classification algorithms
- Rights
- openAccess
- License
- Derechos Reservados - Escuela Colombiana de Ingeniería Julio Garavito, 2018
| Summary: | E-mail communication is still the most prevalent form of Customer support process in Organizations. Therefore, organizations have had to implementing processes focused on email categorization in accordance with email contents, in order to provide an efficient response to customer requests. A common text mining approach involves a representation of text based on keywords later combined with machine learning. This project presents a methodological approach that evaluates classification algorithms: Supoport Vector Machine and Gradient Boodting Trees on a corpus builded from emails database of the Administrative Department of Public Function in Colombia. |
|---|