A Hybrid Approach for Detecting Software Refactoring Sequencing

The process of refactoring improves the software that is created by making changes to the composition of a design, while maintaining the framework of the software. However, addressing these smells without setting suitable priorities may not yield the desired outcomes. Code smells significantly increase maintenance costs and hinder the evolution of systems. To address these issues, refactoring sequencing techniques are utilized to improve a system's design by modifying its internal structure while preserving its outer behavior. However, in huge-scale structure, the variety of code smells that need to be fixed can be substantial, and not all of them can be resolved automatically. Therefore, it becomes crucial to prioritize the order of the code smells according to various criteria such as the risk as well as importance of the affected category. In this paper work, a novel hybrid approach is presented, leveraging the Hybrid Spotted Hyena and Emperor Penguin (HSHEP) optimizer-based algorithm. The primary objective of this approach is to maximize the number of fixed threatful code smells based on the maintainer's opinions and requirements. For evaluating the impact of the proposed approach, it is analyzed on small as well as large-sized open-source systems, targeting five different kinds of code smells.