Application of GPT Language Models for Innovation in Activities in University Teaching
Project Overview
The document explores the integration of generative AI, specifically GPT language models, in higher education, focusing on their transformative role in university teaching. It highlights key applications such as content generation, personalized learning experiences, and support in problem-solving tasks, particularly within Software Engineering courses where ChatGPT was evaluated as a teaching assistant. The findings reveal that while ChatGPT excels in explaining concepts and enhancing engagement, it has limitations in complex problem-solving scenarios. The document also stresses the necessity of addressing ethical considerations surrounding the use of AI in educational settings, advocating for universities to develop clear protocols and guidelines to govern the implementation of these technologies. Overall, the findings suggest that while generative AI holds significant potential to enrich educational practices, careful oversight and ethical frameworks are crucial to ensure its effective and responsible use in academia.
Key Applications
ChatGPT as an assistant for theory, exercises, and laboratory practices in Software Engineering and related courses.
Context: University-level education, specifically in Software Engineering and Development of Emerging Technologies courses for Computer Science and Systems Information Engineering degrees.
Implementation: Utilizing ChatGPT to facilitate student queries in theoretical and practical components of various courses, employing structured activities and evaluation methods to assess its effectiveness across different topics and activities.
Outcomes: Increased understanding of concepts, dynamic learning experiences, and assistance in problem-solving across different courses. Positive evaluations of ChatGPT's utility for explaining concepts, with comparative insights on its effectiveness in different contexts.
Challenges: Challenges include ensuring the integrity of assessments, addressing the potential for misuse of AI in academic work, and variability in ChatGPT's performance based on domain-specific knowledge.
Implementation Barriers
Ethical
The potential for misuse of AI models, leading to academic dishonesty and integrity issues.
Proposed Solutions: Establishing protocols and policies to promote transparency and responsible use of AI in educational settings.
Technical
Variability in the quality of responses from AI models, especially in domain-specific applications.
Proposed Solutions: Conducting evaluations to understand the reliability of responses and refining the use of the model based on feedback.
Project Team
Manuel de Buenaga
Researcher
Francisco Javier Bueno
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Manuel de Buenaga, Francisco Javier Bueno
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