Abstract
Mining Software Repositories in collaborative environment during software evolution or maintenance faces challenges due to creation of larger than necessary slices or unnecessary splitting of Revision History and detection of edge level changes. Due to these limitations, GIT and Diff & Merge Tools does not accurately detect the similarities and changes between versions due to renaming or shifting. Detection of these similarities accurately helps to detect code clones and change patterns that improves understandability, knowledge transfer and tracking changes. Therefore, the authors proposed fuzzy based hybrid technique to detect the similarities/changes between versions considering RS by enriching the Revision History at three granularities- File, Class and Method level. 30% more entities have been found similar/change by deriving Classification model with f-score and ROC Area more than 0.985 and .994 respectively for all applications. Hence, proposed technique improves productivity, reusability and maintainability with respect to VCA.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
