A Novel Hybrid Intelligent Tutoring System and Its Use of Psychological Profiles and Learning Styles

Abstract

This paper presents a novel Intelligent Tutoring System based on traditional and connectionist Artificial Intelligence. It is adaptive and reactive and has the ability to offer customized and dynamic teaching. Features of apprentice’s psychological profile or learning style are employed as basic elements of customization, and they are complemented by (human) expert rules. These rules are represented by probability distributions. The proposed system is implemented on web environment to take advantages such as wide reach and portability. Three types of navigation (on course contents) are compared based on user performances: free (user has full control), random (user is controlled by chance) and intelligent (navigation is controlled by the proposed system: neural network combined with expert rules). Descriptive and inferential analysis of data indicate that the application of proposed techniques is adequate, based on (significant at 5%) results. The main aspects that have been studied are retention (“learning improvement”) normalized gain, navigation total user time and number of steps (length of visited content). Both customizations (by psychological profiles and learning styles) have shown good results and no significant difference has been found between them.