AI CODE REVIEWER AND BUGS PREDICTOR

In conclusion, the development of the AI Code Reviewer and Bug Predictor system provides an effective framework to analyze and ensure the quality and reliability of source code. By systematically evaluating code using artificial intelligence techniques, the system ensures that review processes are accurate, consistent, and result-oriented. The project enhances software development efficiency by allowing developers to examine uploaded code through data-driven methods. The automatic classification of code as buggy or clean helps identify potential errors and supports better decision-making. Furthermore, the use of advanced models such as neural networks ensures that the system can detect complex patterns and hidden issues, improving prediction accuracy. The integration of multiple analysis techniques also provides a comprehensive evaluation of code quality. Overall, this project contributes to improving reliability, transparency, and efficiency in software development. It supports developers in implementing a structured, efficient, and technology-driven approach that helps reduce errors and promotes the development of robust and maintainable software systems.

[1]       T. J. McCabe (1976). A Complexity Measure. IEEE Transactions on Software Engineering.   [2]       S. Panichella, A. Di Sorbo, E. Di Penta (2015). How Developers’ Collaborations Identify Code Defects. IEEE International Conference on Software Maintenance and Evolution (ICSME). [3]       Y. Zhou, H. Leung (2006). Empirical Analysis of Object-Oriented Design Metrics for Predicting High and Low Severity Faults. IEEE Transactions on Software Engineering. [4]       T. Menzies, J. Greenwald, A. Frank (2007). Data Mining Static Code Attributes to Learn Defect Predictors.   IEEE Transactions on Software Engineering.   [5]       V. J. Hellendoorn, P. Devanbu (2017). Deep Learning Type Inference. ACM Joint Meeting on Foundations of Software Engineering (FSE). [6]       M. Allamanis, E. T. Barr, P. Devanbu, C. Sutton (2018). A Survey of Machine Learning for Big Code and Naturalness. ACM Computing Surveys. [7]       Z. Li, D. Zou, S. Xu, H. Jin (2018). VulDeePecker: A Deep Learning-Based System for Vulnerability Detection.   NDSS Symposium.