Multiple coordinated views to support aspect mining using program slicing

Abstract

Aspect Mining and Refactoring to Aspects aim to identify crosscutting concerns and encapsulate them in aspects, respectively. Aspect Mining remains as non-automatic process, i.e., the user needs to analyze and understand the results generated by techniques/tools, and confirm crosscutting concerns to refactor them to aspects. In this paper we propose a visual approach that deals with results generated by two aspect mining techniques proposed in the literature. By coordinating visual mappings, different levels of detail to explore software artifacts support aspect mining facilitating their interpretation for further refactoring to aspects. The model to coordinate multiple views was implemented (SoftV iz4AspectMining tool) and in this paper are presented the visualizations obtained, how to interpret them and lessons learned.