The AI Companion in Education: Analyzing the Pedagogical Potential of ChatGPT in Computer Science and Engineering
Project Overview
The document explores the use of ChatGPT, a generative AI tool, in Computer Science and Engineering education, focusing on its pedagogical potential, key applications, and associated challenges. It highlights how ChatGPT can enhance educational practices by providing personalized learning experiences, assisting with coding tasks, and facilitating problem-solving. However, the analysis also addresses significant concerns related to academic integrity and the reliability of the AI's outputs, which can impact the learning process. To evaluate its effectiveness, a systematic framework for reliability analysis is proposed, aiming to assess and improve the quality of ChatGPT's responses across various educational tasks. Overall, the findings suggest that while generative AI like ChatGPT holds promise for enriching educational experiences, careful consideration of its limitations is essential to ensure it serves as a beneficial tool in the learning environment.
Key Applications
ChatGPT as a virtual assistant for generating educational content, aiding in exam preparation, code generation, debugging, and data analysis.
Context: Used in higher education across various computer science courses, including Artificial Intelligence, Programming, and Data Science. It targets both educators and learners, facilitating learning through practice problems, coding assignments, and data visualization tasks.
Implementation: ChatGPT was employed to generate practice problems, assist in coding assignments and debugging, and provide real-time feedback on programming tasks. Students requested data visualization and analysis tasks, receiving insights and visual outputs from ChatGPT, which were evaluated for effectiveness.
Outcomes: Improved engagement in learning, enhanced critical thinking skills, better exam preparation, higher scores in programming tasks, and improved data handling skills among students.
Challenges: Concerns about academic integrity, reliability of responses, inaccuracies in generated code and visualizations, oversimplification of complex tasks, and the potential decline in critical thinking skills due to reliance on AI.
Implementation Barriers
Academic Integrity
Concerns about students using ChatGPT-generated content without proper understanding or attribution, leading to plagiarism.
Proposed Solutions: Development of clear academic integrity policies that incorporate AI tools and promote ethical use of technology.
Reliability and Skill Development
Inconsistencies and inaccuracies in ChatGPT's responses can mislead learners and impair educational outcomes. Overreliance on ChatGPT may also lead to a decline in students' critical thinking and problem-solving skills.
Proposed Solutions: Establishing a reliability analysis framework to assess and improve the quality of ChatGPT's outputs, while integrating AI tools into curricula and fostering an environment that emphasizes critical analysis and understanding.
Project Team
Zhangying He
Researcher
Thomas Nguyen
Researcher
Tahereh Miari
Researcher
Mehrdad Aliasgari
Researcher
Setareh Rafatirad
Researcher
Hossein Sayadi
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Zhangying He, Thomas Nguyen, Tahereh Miari, Mehrdad Aliasgari, Setareh Rafatirad, Hossein Sayadi
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