Detection of Programming Style Consistency and Plagiarism through Code Similarity Metrics

This study introduces a framework for automatic detection of stylistic and structural similarity in student programming assignments. The approach uses static analysis to extract lexical, syntactic, and stylistic features from C source code. Each submission is represented as a numerical feature vector, which can be compared using cosine similarity, Euclidean distance, and Manhattan distance metrics. The framework includes an interactive component that visualizes similarity results and highlights cases that may require closer inspection. In practice, this approach simplifies the verification of programming style consistency and supports the detection of possible plagiarism in large groups of students. The solution can be easily integrated into existing grading workflows, allowing instructors to quickly focus on the most relevant cases. By making detailed feature comparisons accessible and interactive, the system strengthens transparency in the evaluation process and contributes to higher standards of academic integrity.