A Large Language Model Question Generator Based on Bloom's Taxonomy Template
Hapnes Toba, Laurentius Gusti Ontoseno Panata Yudha, Oscar Karnalim, Hendra Bunyamin, Terutoshi Tada · 2024
Integrating Artificial Intelligence (AI) in education has become a transformative force, particularly in developing teaching materials and automated assessments. This paper explores the potential of AI, precisely OpenAI's natural language processing capabilities, to revolutionize educational practices by automating the generation of diverse and aligned assessment questions. By leveraging AI-driven tools, educators can enhance the scalability and personalization of educational resources, thereby improving student engagement and learning outcomes. This preliminary study evaluates the incorporation of a large language model, ChatGPT-3.5-Turbo, to generate questions for a pre-test activity. In our observation that compares two academic year assessments, the generated pre-test questions could improve the post-test achievement significantly. Our findings suggest that while AI tools like the proposed question generator show the potential to raise concept understanding, there is still room for improvement in challenging students with higher-order thinking skills.