Selección de rasgos a través de los FS-Testores para la representación del modelo del estudiante
Fecha
2007-06-25
Autores
Figuero Maldonado, Darianny
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central “Marta Abreu” de Las Villas
Resumen
El proceso de adquisición del conocimiento dificulta el desarrollo de los Sistemas Basados en el Conocimiento y constituye un obstáculo para la construcción de estos, luego de conocer los rasgos que lo conforman, es conveniente realizar una selección de los rasgos relevantes, siempre que sea posible y a la vez determinar su peso informacional, obteniéndose una simplificación en la representación del conocimiento experto y garantizándose un mejor desempeño computacional. La teoría de los FS-Testores del enfoque Lógico Combinatorio del Reconocimiento de Patrones ha sido utilizada con este fin en el diseño e implementación de Sistemas Basados en Casos de Enseñanza-Aprendizaje con previa selección de rasgos. Se ha desarrollado un módulo que facilita esta tarea como parte de una herramienta que elabora materiales educativos adaptativos, teniendo en cuenta tanto el estado cognitivo como el afectivo del alumno, lográndose una representación eficiente del Modelo del Estudiante dado una simplificación de su modelado inicial.
The development of Knowledge-Based Systems is obstructed by the knowledge acquisition stage, therefore becoming a hindrance. Once the system’s features are known it is convenient, whenever possible, to select the relevant features and, at the same time, to determine their informational weight in order to achieve a more straightforward representation of the expert knowledge and assuring a better computational performance. The FS-Testors theory belongs to the Logical Combinatorial approach of Pattern Recognition and has been used knowingly in the design and implementation of Case-Based Systems of Teaching-Learning, with a previous feature selection. A module has been developed as a part of a tool whose aim is to elaborate educational adaptive materials which take into consideration the student's both cognitive and affective states. This module retrieves an efficient representation of the Student Model owing to a simplification of its initial modeling.
The development of Knowledge-Based Systems is obstructed by the knowledge acquisition stage, therefore becoming a hindrance. Once the system’s features are known it is convenient, whenever possible, to select the relevant features and, at the same time, to determine their informational weight in order to achieve a more straightforward representation of the expert knowledge and assuring a better computational performance. The FS-Testors theory belongs to the Logical Combinatorial approach of Pattern Recognition and has been used knowingly in the design and implementation of Case-Based Systems of Teaching-Learning, with a previous feature selection. A module has been developed as a part of a tool whose aim is to elaborate educational adaptive materials which take into consideration the student's both cognitive and affective states. This module retrieves an efficient representation of the Student Model owing to a simplification of its initial modeling.
Descripción
Palabras clave
FS-Testores, Sistemas Basados en el Conocimiento, Enfoque Lógico Combinatorio, Reconocimiento de Patrones, Modelo del Estudiante, Inteligencia Artificial