🤖 AI Summary
This study addresses the inefficiency of traditional programming exam design and its limited capacity to holistically assess students’ creativity, problem-solving skills, and domain knowledge. It presents the first systematic application of prompt engineering to the automatic generation of programming examination questions, proposing a method that leverages carefully crafted, diverse prompt templates to guide ChatGPT—without requiring fine-tuning of large language models. The approach autonomously produces high-quality questions and reference solutions spanning theoretical and practical aspects, multiple question types, and varying difficulty levels. Experimental results demonstrate that the generated items match or exceed the quality of human-authored questions while substantially improving item development efficiency. User studies further confirm the method’s effectiveness and practical value in educational settings.
📝 Abstract
In computer science, students are encouraged to learn various programming languages such as Python, C++, and Java, equipping them with a broad range of technical skills and problem-solving capabilities. Nevertheless, the design of objective examination questions to assess students' creativity, problem-solving abilities, and domain knowledge remains a significant challenge. This paper proposes a methodology to address these challenges by leveraging prompt engineering techniques with ChatGPT. Prompt engineering is an efficient technique that optimizes the performance of language models, enabling the automatic generation of high-quality exam questions with varying types and difficulty levels, all without requiring additional fine-tuning of the model. This study applies diverse patterns and templates to generate exam questions that incorporate both theoretical and practical components, thereby facilitating a comprehensive evaluation of students' theoretical understanding and hands-on programming proficiency. A survey was conducted to validate the proposed method, and although certain areas indicated room for improvement, the overall results confirmed its significance and relevance. The generated questions and model answers exhibit quality comparable to, or even surpassing, manually crafted questions while significantly reducing the time and effort required for question preparation. This research demonstrates that automated exam question generation through prompt engineering enhances the quality and efficiency of assessment tools in education, establishing it as a valuable asset for future educational environments.