Designing an Agent-based Intelligent Teaching Assistant and Evaluating its Efficiency in e-Learning Portals

Abstract

Introduction: Today, several educational portals are established by organizations to enhance learning. The use of intelligent agents is necessary to improve a system's quality and cover limitations such as face-to-face relation. Focusing on human-based agents, the researchers designed an intelligent agent as an educational assistant and then its efficiency was evaluated. Some applicable suggestions for developing intelligent agents as educational assistants are provided to virtual universities.Materials and Methods: This study was carried out in two steps: prototyping an intelligent agent and then evaluating it in three educational groups (MBA, Technology Management, and Industrial Management) at Mehr Alborz Institute in 2013. The participants consisted of 31 Electronic students selected based on random sampling. A questionnaire was used as the research tool with a reliability of 85%, estimated by Cronbach’s alpha coefficient. Both descriptive and inferential statistical procedures were used to analyze the data, using SPSS software, version 21.Results: By examining the results of the interaction of 31 students with the designed agent through Mehralborz institute learning management system, and the conceptual model based on e-learning effectiveness, user friendliness, user satisfaction, and efficiency variables gained the scores of 55, 58 and 57, respectively with the significance levels of 0.03, 0.09, and 0.01. This shows that this agent was overall moderately efficient.Conclusion: The findings revealed the effectiveness of the agent-based intelligent teaching assistant in e-learning portals.