Our systematic review of source code authorship attribution is now published in Computers (MDPI, open access). It is joint work with Bilal Al-Ahmad, Murad Magableh, Sahar Mahmoud, Doaa Altayeb, and Dana Dghaym.

About the paper

Source code authorship attribution is the task of identifying the programmer who wrote a piece of code. It is useful for plagiarism detection, forensic investigation of malware, and software provenance. The review organizes the published research into four stages of the attribution pipeline: preprocessing, feature extraction, classification, and prediction. It covers stylometric metrics, n-grams, abstract syntax trees, program graphs, learned embeddings, and watermarking. We compare reported results across C, C++, C#, Java, and Python, with attention to dataset size and evaluation setting. We close with open questions on cross-language and open-set attribution, robustness to code transformation and mimicry, AI-assisted and collaborative code, and interpretable forensic evidence.

DOI: https://doi.org/10.3390/computers15110843

Article page: MDPI

PDF: Published version

Citation

Al-Ahmad, B., Harshvardhan, M., Magableh, M., Mahmoud, S., Altayeb, D., & Dghaym, D. (2026). Source Code Authorship Attribution: Trends, Techniques, and Challenges. Computers, 15(11), 843. https://doi.org/10.3390/computers15110843