Survey and analysis of quality measures used in aspect mining

Aspect mining investigates effective ways of finding crosscutting concerns in existing non-aspect oriented software. These crosscutting concerns can then be refactored into aspects to reduce the system's complexity and make it easier to understand, maintain, and evolve. There have been numerous studies introducing different aspect mining techniques, but they used different quality measures to evaluate their techniques. This paper consolidates a list of these existing quality measurements, discusses how they differ from each other, identifies some examples of how they have been used in previous aspect mining studies, and conducts an analysis of the commonly used metrics for aspect mining clustering. The metrics are compared using real and sample clustering results, identifying their similarities and differences, as well as their strengths and weakness.