Personalized Programming Guidance based on Deep Programming Learning Style Capturing
Project Overview
The document presents the Programming Exercise Recommender with Learning Style (PERS), an innovative model designed to enhance programming education through personalized guidance tailored to individual learners' intrinsic programming behaviors and learning styles. By leveraging techniques from sequential recommendation and integrating the Felder-Silverman Learning Style Model, PERS effectively adapts to the unique needs of each student. The findings from extensive experiments demonstrate that PERS significantly improves learner engagement and reduces dropout rates in programming courses, highlighting its potential as a transformative tool in educational settings. Overall, the application of generative AI in education, as illustrated by the PERS model, showcases the ability to create customized learning experiences that foster academic success and retention among students.
Key Applications
Programming Exercise Recommender with Learning Style (PERS)
Context: Online programming education for college students
Implementation: PERS was developed to simulate learners' intricate programming behaviors and provide personalized recommendations based on their learning styles.
Outcomes: Improved learning engagement, reduced dropout rates, and enhanced personalized guidance in programming education.
Challenges: Challenges in recognizing complex programming behaviors and capturing intrinsic learning patterns.
Implementation Barriers
Technical
Difficulty in recognizing complex programming behaviors and capturing intrinsic learning patterns.
Proposed Solutions: The proposed model incorporates a differentiating module to address these challenges by capturing fine-grained learning patterns.
Pedagogical
Need for alignment of the model with actual learning processes and preferences.
Proposed Solutions: Incorporation of pedagogical theories such as the Felder-Silverman Learning Style Model to enhance interpretability and alignment.
Project Team
Yingfan Liu
Researcher
Renyu Zhu
Researcher
Ming Gao
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Yingfan Liu, Renyu Zhu, Ming Gao
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