Constructing Object Groups Corresponding to Concepts for Recovery of a Summarized Sequence Diagram

03/06/2020 ∙ by Kunihiro Noda, et al. ∙ 0

Comprehending the behavior of an object-oriented system solely from its source code is troublesome, owing to its dynamism. To aid comprehension, visualizing program behavior through reverse-engineered sequence diagrams from execution traces is a promising approach. However, because of the massiveness of traces, recovered diagrams tend to become very large, causing scalability issues. To address the issues, we propose an object grouping technique that horizontally summarizes a reverse-engineered sequence diagram. Our technique constructs object groups based on Pree's meta patterns, in which each group corresponds to a concept in the domain of a subject system. Visualizing interactions only among important groups, we generate a summarized sequence diagram depicting a behavioral overview of the system. Our experiment showed that our technique outperformed the state-of-the-art trace summarization technique in terms of reducing the horizontal size of reverse-engineered sequence diagrams. Regarding the quality of object grouping, our technique achieved an F-score of 0.670 and a Recall of 0.793 on average under the condition of #lifelines (i.e., the horizontal size of a sequence diagram) < 30, whereas those of the state-of-the-art technique were 0.421 and 0.670, respectively. The runtime overhead imposed by our technique was 129.2 average, which is relatively smaller in the literature.



There are no comments yet.


page 2

page 3

page 5

page 8

page 9

page 11

page 13

page 15

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.