Code Smell

Code smells represent suboptimal design choices in software, hindering maintainability and reusability. Current research focuses on automated detection of code smells using machine learning techniques, including federated learning and neural word embeddings, across various contexts such as Python code generated by AI tools and machine learning applications themselves. This work aims to improve software quality and reduce technical debt by identifying and addressing these issues early in the development lifecycle, impacting both software engineering practices and the reliability of AI systems.