AI-Driven Interface Design for Intelligent Tutoring System Improves Student Engagement
Project Overview
The document explores the role of generative AI in education, particularly through its application in Intelligent Tutoring Systems (ITS) to boost student engagement and learning outcomes. It presents findings from an A/B testing study conducted on a mobile ITS named Santa, designed to help users prepare for the TOEIC exam. The implementation of AI-driven interface enhancements led to notable improvements in key user engagement metrics, including conversion rates and average revenue per user. The results indicate that thoughtfully designed AI interfaces not only elevate student motivation but also significantly enhance the learning experience, showcasing the potential of generative AI to transform educational tools and foster better academic performance. This research underscores the effectiveness of integrating advanced AI features into educational platforms to facilitate personalized learning and improve overall educational efficacy.
Key Applications
AI-driven interface design for Intelligent Tutoring System (ITS) named Santa
Context: Educational context focusing on TOEIC test preparation for users, particularly students seeking to improve their English proficiency.
Implementation: The interface was designed to provide personalized feedback and analytics based on users' diagnostic test results, using AI components to improve engagement.
Outcomes: Controlled A/B tests showed improvements in engagement metrics by up to 25.13% with the new interface design.
Challenges: Challenges include the need for a well-designed user interface that effectively conveys AI analysis and the potential for outdated methods in interface design.
Implementation Barriers
Technical Barrier
The interface of ITS may not fully support transparency of AI's analysis to students, which can negatively affect engagement. There is also a need for improved interface design to better utilize AI features and increase explainability of AI-driven analytics.
Proposed Solutions: Improving the interface design to enhance transparency, engagement, and explainability of AI-driven analytics.
Research Barrier
Most studies in ITS have focused on knowledge diagnosis and learning item suggestion rather than user interface design. There is a need for increased focus on user interface development and empirical evaluations of their effectiveness in engaging students.
Proposed Solutions: Increased focus on user interface development and empirical evaluations of their effectiveness in engaging students.
Project Team
Byungsoo Kim
Researcher
Hongseok Suh
Researcher
Jaewe Heo
Researcher
Youngduck Choi
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Byungsoo Kim, Hongseok Suh, Jaewe Heo, Youngduck Choi
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai