Skip to main content Skip to navigation

Choose Your Own Question: Encouraging Self-Personalization in Learning Path Construction

Project Overview

The document explores the innovative application of generative AI in education through the development of Rocket, a user-friendly interface designed to enhance Interactive Educational Systems (IESs). Rocket addresses existing limitations in IESs, such as the lack of student feedback mechanisms and transparency in content recommendations. By facilitating self-personalization, Rocket empowers students to curate their learning experiences, allowing them to select materials that align with their preferences and learning styles. The system leverages AI to extract insights and visually represent students' learning progress, thereby promoting engagement and explainability. The findings suggest that such advancements not only improve student satisfaction and involvement but also foster a more tailored educational journey, ultimately enhancing learning outcomes.

Key Applications

Rocket - a Tinder-like User Interface for learning material selection

Context: Interactive Educational Systems (IES) targeting students seeking personalized learning paths.

Implementation: Implemented as a mobile touchscreen or web application allowing students to swipe or tap to choose learning materials.

Outcomes: Improved student engagement, enhanced explainability of recommendations, and better tracking of learning progress.

Challenges: Ensuring the system accommodates diverse learning preferences and maintaining student interest over time.

Implementation Barriers

Technical Barrier

Limited innovation in User Interfaces of existing IESs, leading to a lack of engagement and transparency. Additionally, students have limited insight into the recommendation process, which can reduce trust in the IES.

Proposed Solutions: Development of Rocket that includes visual summaries and allows for self-personalization based on student choices. Rocket enhances transparency by showing AI-extracted features used in recommendations.

Project Team

Youngduck Choi

Researcher

Yoonho Na

Researcher

Youngjik Yoon

Researcher

Jonghun Shin

Researcher

Chan Bae

Researcher

Hongseok Suh

Researcher

Byungsoo Kim

Researcher

Jaewe Heo

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Youngduck Choi, Yoonho Na, Youngjik Yoon, Jonghun Shin, Chan Bae, Hongseok Suh, Byungsoo Kim, Jaewe Heo

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

Let us know you agree to cookies