Development Of A Tool That Uses Machine Learning To Review Ant) Provide Feedback On Code Quality:- Uwamacha Chidinma H

ABSTRACT

 The increasing complexity of software development necessitates robust methods for ensuring code qunliiy, as poor-quality code can lead to significant issues such as bugs, security vulnerabilities, and

high maintenance costs. This project aims to develop a tool that leverages machine learning (ML) techniques to automate the assessment of code quality, thereby providing developers with timely and objective feedback. The study begins with an exploration of traditional code review processes, highlighting their limitations, particularly in terms of consistency and scalability. Key metrics for evaluating code quality, such as complexity, readability, and maintainability, are identified and

incorporated into the tool's design. Utilizing datasets from open-source projects, a machine learning model is trained to analyze code and detect common issues, such as code smells and anti-patterns. The effectiveness of the tool is then evaluated against conventional review methods, focusing on metrics such as accuracy and user satisfaction. Results indicate that the ML-based tool significantly enhances code review processes, reducing time and improving consistency. This study underscores the potential of integrating machine learning into software engineering practices, offering a scalable solution for mainlaining high code quality standards. Future work will focus on refining the model, expanding lcinguage support, and further integrating the tool into existing development workfows.