Ingeniería del conocimiento automatizada en la creación del modelo del estudiante de los sistemas de enseñanza-aprendizaje inteligentes
Archivos
Fecha
2007-05-25
Autores
León Espinosa, Maikel
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central “Marta Abreu” de Las Villas
Resumen
Las Tecnologías de la Información y las Comunicaciones han permitido la evolución de la enseñanza asistida por computadoras, y el uso de técnicas de Inteligencia Artificial ha favorecido eldesarrollo cualitativo de los Sistemas de Enseñanza-Aprendizaje que tienen en cuenta la de los estudiantes. Sin embargo, el proceso de adquisición del conocimiento continúa siendo una etapa difícil y necesaria al implementar este tipo de sistemas. En este trabajo se desarrolla una herramienta computacional que permite crear de forma automatizada la Base de
Conocimiento que representa al alumno en un Sistema de Enseñanza-Aprendizaje Inteligente. La arquitectura de la herramienta la integran varios módulos de ayuda a la importante y difícil tarea de ingeniería del conocimiento, permitiéndose de esta forma la obtención de rasgos cognitivos y afectivos-motivacionales que caracterizan al alumno. El profesor elabora las preguntas interactuando con un editor dinámico, a partir de las cuales se obtienen los rasgos cognitivos. Por su parte, los
rasgos afectivos-motivacionales son captados a través de un sistema que permite la confección, perfeccionamiento y aplicación de test con un estilo conversacional. Para lograr una comparación entre los términos y las clasificaciones del test se hace uso de un Sistema Basado en Casos y de la aplicación de técnicas para la comparación de textos en el procesamiento de Lenguaje Natural.
Luego de conocer los rasgos que conforman el modelado inicial del alumno, se aplican técnicas de selección de rasgos, utilizándose los FS-Testores, basados en el enfoque Lógico Combinatorio del Reconocimiento de Patrones, para simplificar la representación inicial del conocimiento y garantizar un mejor desempeño computacional de los Sistemas de Enseñanza-Aprendizaje Inteligentes. Se
logra entonces una herramienta que permite la transferencia automatizada del conocimiento de los profesores a bases de conocimiento con fines educativos, donde los expertos construyen una representación simplificada del Modelo del Estudiante con resultados satisfactorios.
The Information & Communication Technologies have allowed the evolution of the computer-aided education whereas the employment of Artificial Intelligence techniques has forwarded the qualitative development of the Teaching-Learning Systems which take into account the features of each student. Yet the knowledge acquisition process remains a tough albeit necessary stage at the time of building such sort of systems. A computational tool for the automated creation of the knowledge base representing a student in an Intelligent Teaching-Learning System is envisioned and deployed in this study. Its architecture is comprised of several modules which aid to the difficult and troublesome task of getting knowledge from the human expert, thus enabling the obtaining of the cognitive as well as affective-motivational features which characterize the students. The instructor works out the questions by means of the interaction with a dynamic editor. From such questions, the cognitive features are derived. On the other hand, the affective-motivational features are captured through a system which allows the makeup, enhancement and application of conversational-styled tests. In order to establish a comparison between the tests’ terms and classifications, a Case-Based System and several text comparison procedures borrowed from Natural Language Processing are utilized. Once the features comprising the initial modeling of the student are known, feature selection techniques come into play, especially suggesting the use of FS-Testors, which fully rely on the Logical Combinatory approach of Pattern Recognition so as to simplify the knowledge’s early representation and guarantee a better computational performance of the Intelligent Teaching- Learning Systems. The final product is a tool which permits the automated transfer of expert knowledge flowing from instructors to knowledge bases with educational ends, wherein the very experts build their own abbreviated representation of the Student Model with satisfactory results.
The Information & Communication Technologies have allowed the evolution of the computer-aided education whereas the employment of Artificial Intelligence techniques has forwarded the qualitative development of the Teaching-Learning Systems which take into account the features of each student. Yet the knowledge acquisition process remains a tough albeit necessary stage at the time of building such sort of systems. A computational tool for the automated creation of the knowledge base representing a student in an Intelligent Teaching-Learning System is envisioned and deployed in this study. Its architecture is comprised of several modules which aid to the difficult and troublesome task of getting knowledge from the human expert, thus enabling the obtaining of the cognitive as well as affective-motivational features which characterize the students. The instructor works out the questions by means of the interaction with a dynamic editor. From such questions, the cognitive features are derived. On the other hand, the affective-motivational features are captured through a system which allows the makeup, enhancement and application of conversational-styled tests. In order to establish a comparison between the tests’ terms and classifications, a Case-Based System and several text comparison procedures borrowed from Natural Language Processing are utilized. Once the features comprising the initial modeling of the student are known, feature selection techniques come into play, especially suggesting the use of FS-Testors, which fully rely on the Logical Combinatory approach of Pattern Recognition so as to simplify the knowledge’s early representation and guarantee a better computational performance of the Intelligent Teaching- Learning Systems. The final product is a tool which permits the automated transfer of expert knowledge flowing from instructors to knowledge bases with educational ends, wherein the very experts build their own abbreviated representation of the Student Model with satisfactory results.
Descripción
Palabras clave
Ingeniería del Conocimiento Automatizada, Modelo del Estudiante, Rasgos Cognitivos, Rasgos Afectivos-Motivacionales, Sistemas de Enseñanza-Aprendizaje Inteligentes