DIGITAL LIBRARY
DEVELOPMENT OF AN INTELLIGENT TUTORING SYSTEM OF GENERALIZED SUPPORT FOR DIFFERENTIATED LEARNING
1 Instituto Tecnológico Autónomo de México (MEXICO)
2 Universidad Nacional Autónoma de México (MEXICO)
About this paper:
Appears in: EDULEARN19 Proceedings
Publication year: 2019
Pages: 10145-10151
ISBN: 978-84-09-12031-4
ISSN: 2340-1117
doi: 10.21125/edulearn.2019.2538
Conference name: 11th International Conference on Education and New Learning Technologies
Dates: 1-3 July, 2019
Location: Palma, Spain
Abstract:
In this work we present a computational system for learning support called SAGE (Sistema de Apoyo Generalizado para la Enseñanza Individualizada) designed to offer a teaching plan for each student according to their skills and knowledge based on a taxonomy of learning objectives. To achieve it, a content map and the Bloom’s Taxonomy were used. The content map organizes subjects from the general to the particular through a Morganov-Heredia matrix where dependencies are established and the existent relationships between the course subjects are represented. The model of skills and knowledge is based on the first four cognitive levels from Bloom’s Taxonomy and it is obtained from a diagnostic test and updated according to the advance of the students. The system consists of three modules that were created according to the object-oriented methodology: the student module, the teacher module and the interface module. The student module takes the lessons and consults their evaluations, the teacher model registers students to the course and follows up on their progress and the interface module provides a simple interaction with the users, keeping the student’s attention during the lessons and facilitating the query of information to the teacher. Our final system is content-free, integrates some support tools like games and practice exercises and allows students and teachers to check the progress of the course by comparing the scores with the group average, showing positions inside the group, median and standard deviations, as well as charts that show progress from one chapter to another. We also propose three improvements for the system: a clustering analysis to the cognitive characteristics of the students for determining grouping profiles, the Bayesian Knowledge Tracing method based on those profiles so that progress depends on the probabilities of students having the required knowledge but failing in the test or not having the knowledge and guessing the answers and the substitution of Bloom’s Taxonomy by Marzano’s Taxonomy since the last one takes into account important aspects as metacognition as a first approach to learning.
Keywords:
Intelligent Tutoring Systems, Bloom’s Taxonomy, Individual Teaching, Marzano’s Taxonomy, Bayesian Knowledge Tracing.