Collaborative and AI-aided Exam Question Generation using Wikidata in Education
Project Overview
The document explores the innovative use of generative AI in education through the development of 'PhysWikiQuiz', a system designed to automate the creation of exam questions for physics students by utilizing Wikidata. It highlights the difficulties educators encounter in crafting personalized questions and presents a collaborative framework for question generation that alleviates the burden on teachers. By integrating a Computer Algebra System, the tool not only generates diverse and individualized questions but also verifies the correctness of answers, ultimately aiming to improve student learning outcomes while significantly reducing the workload for educators. The findings suggest that such AI applications can enhance educational efficiency, foster deeper student engagement, and streamline the assessment process in physics education, showcasing the potential of generative AI to transform traditional teaching methodologies.
Key Applications
PhysWikiQuiz
Context: Physics education for teachers and students, especially in a digital learning environment.
Implementation: The system uses Wikidata to retrieve physics knowledge, generates questions based on input formulas, and verifies student answers using a Computer Algebra System.
Outcomes: The system can generate an unlimited number of questions tailored to individual students, enhancing personalized learning and potentially improving student performance.
Challenges: Challenges include accuracy in retrieving and translating formula information, ensuring correct substitutions for identifiers, and generating explanation texts.
Implementation Barriers
Technical
Issues with the accuracy of retrieving formula semantics and units from Wikidata, leading to incorrect question generation.
Proposed Solutions: Improving data quality in Wikidata and refining the formula translation process through community feedback.
Educational
Teachers may lack familiarity with AI tools, affecting their willingness to adopt the system.
Proposed Solutions: Providing training and support for teachers to integrate AI tools into their teaching practices.
Project Team
Philipp Scharpf
Researcher
Moritz Schubotz
Researcher
Andreas Spitz
Researcher
Andre Greiner-Petter
Researcher
Bela Gipp
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Philipp Scharpf, Moritz Schubotz, Andreas Spitz, Andre Greiner-Petter, Bela Gipp
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