Teaching artificial intelligence through drug-drug interaction clustering analysis: Integrating project-based...

Area of Science:

• Educational Technology
• Artificial Intelligence in Education
• Computational Science

Background:

• Artificial intelligence (AI) is increasingly integrated into daily life and scientific research.
• Existing AI education often targets computer science majors, limiting access for undergraduates from non-computing backgrounds.
• A need exists for accessible AI instruction for early-stage undergraduates from diverse academic fields.

Purpose of the Study:

• To develop and evaluate an AI course for non-computing undergraduates.
• To integrate project-based learning (PBL) with large language models (LLMs) for AI education.
• To provide a framework for using LLMs as assistive tools in an undergraduate AI curriculum.

Main Methods:

• Designed a project-based AI course centered on drug-drug interaction network clustering analysis.
• Developed four progressive assignments utilizing LLMs to support programming, data analysis, and interpretation.
• Students engaged in a full data science workflow without requiring prior pharmacology or programming knowledge.

Main Results:

• Preliminary feedback indicates the course enhances problem-solving abilities.
• The project-based approach with LLM integration significantly increases student engagement.
• The curriculum successfully supports students through data curation, algorithm implementation, and result evaluation.

Conclusions:

• The study presents a viable framework for integrating LLMs into PBL for AI education.
• This teaching model is valuable for educators aiming to create or enhance AI courses for diverse student populations.
• The approach demonstrates the potential of LLMs to democratize AI education and foster interdisciplinary learning.