earticle

영어

We introduce an automated method for generating data structure questions using the SMT Solver 'Z3'. Recognizing challenges in traditional question-setting, which might contain logical errors or lack originality, we focus on generating logically consistent and diverse questions. We harness Z3's EnumSort() and Const() functions for conceptual question creation. For applied questions, three main approaches were employed: generating data structures with specific constraints, deriving accurate solutions for questions, and assessing data structure integrity pre and post specific code execution. The outcome not only refines problem creation but also provides a quality-assured learning platform. By transitioning our findings to a web service, we envision bridging educational gaps caused by location and financial constraints. This initiative offers a novel perspective in the realm of educational assessments.