Context-Sensitive Identification of Refactoring Opportunities

Refactoring is a popular procedure for improving the internal structure of software systems. Refactoring is widely practiced by developers, and considerable development effort has been invested in refactoring tooling support. The identification of refactoring opportunities usually is based on subjective perceptions. At best, identification of refactoring opportunities is based on observation of code smells, which is also subjective and error-prone. As a consequence, developers struggle to identify when large systems should be refactored and, once the identification is complete, they still need to choose the types of refactoring to be performed. To overcome these problems, this PhD research aims at: (i) improving the state-of-the-art by introducing a novel model for identifying refactoring opportunities; and (ii) indicating which types of refactorings should be performed in order to improve the code internal structure.