ALGORITMOS EVOLUTIVOS DE MINERÍA DE DATOS DESCRIPTIVA PARA FLUJOS CONTINUOS DE DATOS
Fecha
2019-09-06
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Jaén: Universidad de Jaén
Resumen
En este trabajo se presenta un enfoque para la extracción de patrones emergentes en flujos de datos. Procesa las instancias por bloques siguiendo un enfoque de reentrenamiento. El algoritmo de aprendizaje es un sistema evolutivo difuso en el que se emplean conocimientos previos para adaptarse al cambio de concepto. Se ha realizado un amplio estudio experimental para demostrar tanto la idoneidad del enfoque en la lucha contra el concepto como la calidad de los conocimientos extraídos. Finalmente, la propuesta se aplica a un estudio real relacionado con la determinación continua de los perfiles de los clientes de taxis de la ciudad de Nueva York en función de su tarifa, con el fin de mostrar su potencial.
In this work, an approach for the extraction of emerging patterns in data streams is presented. It processes the instances by means of batches following a retraining approach. The learning algorithm is an evolutionary fuzzy system where previous knowledge is employed in order to adapt to concept drift. A wide experimental study has been performed in order to show both the suitability of the approach in combating concept drift and the quality of the knowledge extracted. Finally, the proposal is applied to a case study related to the continuous determination of the profiles of New York City cab customers according to their fare amount, in order to show its potential.
In this work, an approach for the extraction of emerging patterns in data streams is presented. It processes the instances by means of batches following a retraining approach. The learning algorithm is an evolutionary fuzzy system where previous knowledge is employed in order to adapt to concept drift. A wide experimental study has been performed in order to show both the suitability of the approach in combating concept drift and the quality of the knowledge extracted. Finally, the proposal is applied to a case study related to the continuous determination of the profiles of New York City cab customers according to their fare amount, in order to show its potential.