Hierarchical Clustering for Identifying Crosscutting Concerns in Object Oriented Software Systems

Crosscutting concerns are parts of a program that affect or crosscut other concerns. Usuallythese concerns cannot be cleanly decomposed from the rest of the system, and they are mixed with manycore concerns from the system leading to code scattering and code tangling, and, also, to systems thatare hard to explore and understand. Identifying crosscutting concerns automatically improves both themaintainability and the evolution of the software systems. Aspect mining is a research direction thattries to identify crosscutting concerns in already developed software systems, without using the aspectoriented paradigm. The goal is to identify them and then to refactor them to aspects, to obtain a systemthat can be easily understood, maintained and modified. In this paper we are focusing on the problem ofidentifying crosscutting concerns in object oriented software systems using a hierarchical agglomerativeclustering approach. We experimentally validate our approach on the open source case study JHotDrawand on a real software system. A comparison of our approach with similar existing work is also provided