Prioritisation of code clones using a genetic algorithm

Context: Code clones are prevalent, and due to their diverse impact on projects’ quality they require a proper management strategy. Objectives: Develop GA-based Refactoring-Aware Detection (RAD) approach for prioritisation of code clones. Method: A genetic algorithm (GA) that balances estimated gain and cost/risk of refactoring to select the optimal clone candidate to refactor. Results: GA converges on a solution, with diverse variance. The value of fitness function is higher for multi-objective approaches, but they also exhibit higher variance. Conclusion: GA can be effectively applied for clone prioritising.