Smart Code Review Assistant with Natural Language Processing (Nlp)
Scientia Africana · 29.04.2026 · 25 (2) · 315-328
DOI: 10.4314/sa.v25i2.29 ↗Kurz gefasst
Fachartikel in der Zeitschrift Scientia Africana, online erschienen am 29.04.2026. Verfasst von E. S. Mughele und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Code review is a vital part of the software development lifecycle, promoting better code quality, knowledge sharing, and defect detection. However, manual code reviews are often time-consuming, inconsistent, and subject to human error. This paper presents a Smart Code Review Assistant (SCRA) powered by Natural Language Processing (NLP) techniques and static code analysis to augment the code review process. The system leverages spaCy for NLP tasks including tokenization, part-of-speech tagging, lemmatization, and dependency parsing and Python's AST module for structural code analysis. The assistant detects code smells, logical flaws, and documentation gaps while delivering intelligible, context-aware suggestions for both novice and experienced developers. The system was rigorously evaluated against 80 annotated Python code snippets, achieving a Precision of 89%, Recall of 93%, and an F1-Score of 91%. In addition, a user study involving 32 participants demonstrated an overall satisfaction score of 4.41/5.0 and a 91% satisfaction rate (rating ≥4). Comparative analysis against SonarQube, Amazon CodeGuru, and DeepCode further validates SCRA's competitiveness, particularly in lightweight, low-resource deployment contexts.
Zitieren
APA
Mughele, E. S., Akinola, O. S., & Egbon, C. C. (2026). Smart Code Review Assistant with Natural Language Processing (Nlp). Scientia Africana, 25(2), 315-328. https://doi.org/10.4314/sa.v25i2.29
Vancouver
Mughele ES, Akinola OS, Egbon CC. Smart Code Review Assistant with Natural Language Processing (Nlp). Scientia Africana. 2026;25(2):315-328. doi:10.4314/sa.v25i2.29
BibTeX
@article{Mughele2026,
author = {Mughele, E. S. and Akinola, O. S. and Egbon, C. C.},
title = {Smart Code Review Assistant with Natural Language Processing (Nlp)},
journal = {Scientia Africana},
year = {2026},
volume = {25},
number = {2},
pages = {315--328},
doi = {10.4314/sa.v25i2.29}
}
EUVIMED