Condensing reverse engineered class diagrams through class name based abstraction

In this paper, we report on a machine learning approach to condensing class diagrams. The goal of the algorithm is to learn to identify what classes are most relevant to include in the diagram, as opposed to full reverse engineering of all classes. This paper focuses on building a classifier that is based on the names of classes in addition to design metrics, and we compare to earlier work that is based on design metrics only. We assess our condensation method by comparing our condensed class diagrams to class diagrams that were made during the original forward design. Our results show that combining text metrics with design metrics leads to modest improvements over using design metrics only. On average, the improvement reaches 5.3%. 7 out of 10 evaluated case studies show improvement ranges from 1% to 22%.

I. Introduction

The goal of reverse engineering is to recover knowledge about a software system, in support of a specific engineering task [20]. However, current reverse engineering techniques suffer from several problems, one of them is that the representations they generate are typically complete. Although most of the commercial Computer Aided Software Engineering (CASE) tools are able to leave out some properties in a class diagram, the results are still hard for developers to comprehend because it is not easy to identify which classes are less or more important. A controlled experiment conducted by Fernandez-Saez et al. [4] found that many subjects did not consider reverse engineered diagrams to be helpful.