This study focuses on enhancing an automated SQL statement grading model by expanding the dataset and integrating Local Interpretable Model-agnostic Explanations (LIME) to improve explainability. By incorporating a significantly larger dataset, the model’s ability to generalize across a variety of SQL queries has been enhanced, as demonstrated by improved performance metrics such as precision, recall, and F1 scores. The integration of LIME provides insights into the inference processes of the model, highlighting the influence of specific SQL components on assessment outcomes. These enhancements have practical implications, including more accessibility for users to understand the rationale behind model decisions, potentially leading to more effective learning experiences.