Skip to main content Skip to navigation

EDBooks: AI-Enhanced Interactive Narratives for Programming Education

Project Overview

The document explores the innovative use of generative AI in education through the development and evaluation of EDBook, a platform that combines Large Language Models (LLMs) with conventional learning resources to enhance programming education. EDBook aims to foster personalized learning experiences by facilitating dialogic interactions, enabling students to ask questions and receive customized responses while adhering to a structured learning trajectory. The findings emphasize EDBook's effectiveness in boosting student engagement and promoting active learning, as well as its capability to assess understanding through interactive content. Overall, the integration of generative AI via EDBook demonstrates significant potential for transforming educational practices by tailoring learning experiences to individual needs, thereby improving outcomes in programming education.

Key Applications

EDBook: AI-Enhanced Interactive Narratives for Programming Education

Context: Asynchronous programming education for learners, particularly in programming courses.

Implementation: EDBook combines goal-oriented dialog trees with open-ended LLM interactions to provide structured yet flexible learning experiences.

Outcomes: Increased engagement, personalized learning experiences, and improved assessment scores. Participants spent more time learning and attempted more coding quizzes.

Challenges: Balancing open-ended exploration with structured learning goals; ensuring LLM responses are accurate and pedagogically sound.

Implementation Barriers

Technical Barrier

Difficulty in ensuring that LLM responses align with educational goals and are accurate.

Proposed Solutions: Implementing mechanisms to constrain LLM outputs and using dialog trees to guide learning.

Engagement Barrier

Potential lack of critical thinking due to reliance on pre-written responses.

Proposed Solutions: Encouraging more user-driven exploration and interaction with the content.

Navigational Barrier

Navigating content can be challenging due to the dialog format and less refined search functionalities compared to standard web interfaces.

Proposed Solutions: Improving navigability through more concise content and enhanced search features.

Project Team

Steve Oney

Researcher

Yue Shen

Researcher

Fei Wu

Researcher

Young Suh Hong

Researcher

Ziang Wang

Researcher

Yamini Khajekar

Researcher

Jiacheng Zhang

Researcher

April Yi Wang

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Steve Oney, Yue Shen, Fei Wu, Young Suh Hong, Ziang Wang, Yamini Khajekar, Jiacheng Zhang, April Yi Wang

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