Publication

Abstract : Automated assessment of student code remains a significant challenge in introductory programming education. To address this, we propose a Structure-Based Code Assessment Model (SBCAM), leveraging tools that support Abstract Syntax Tree (AST) analysis techniques, along with innovative feature extraction methodologies. Our model streamlines code assessment by acquiring source code directly from instructors and transforming it into standardized templates. These templates serve as a benchmark for evaluating student submissions, and ensuring adherence to assignment specifications. By automating the assessment process of student submissions using instructor solution, SBCAM reduces manual grading time and provides instructors with valuable insights through comprehensive scoring mechanisms. Ultimately, SBCAM aims to enhance efficiency, accuracy, and instructor productivity, while empowering students to improve their coding skills effectively from the ground up.