Toward experimental evaluation of subsystem classification recovery techniques
Authors:
- A. LakhotiaALA. Lakhotia
- J.M. GravleyJGJ.M. Gravley
- 19 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 262-269
Abstract
Several reverse engineering techniques classify software system components into subsystems. These techniques are designed to discover such classifications when the classifications are unknown. The techniques are rested and evaluated, however, by matching the classifications they recover against expected classifications. Several such techniques may be compared by experimentally evaluating their performance on the same set of software systems. Two things are needed to ensure experiment repeatability: a set of "real-world" software systems whose expected subsystem classifications are known; and an objective criterion to quantitatively determine the similarity of subsystem classifications. This paper contributes to both needs by identifying a set of widely used and easily accessible software systems whose modular decomposition either is documented or can be easily inferred from their design philosophy, and by presenting a measure to quantitatively determine the congruence between hierarchical subsystem classifications.Keywords
This publication has 15 references indexed in Scilit:
- Composing subsystem structures using (k,2)-partite graphs2002
- An object finder for program structure understanding in software maintenanceJournal of Software Maintenance: Research and Practice, 1994
- A new approach to finding objects in programsJournal of Software Maintenance: Research and Practice, 1994
- An Integrated Approach to Software Engineering1991
- Extracting and restructuring the design of large systemsIEEE Software, 1990
- Integrating noninterfering versions of programsACM Transactions on Programming Languages and Systems, 1989
- The TAME project: towards improvement-oriented software environmentsIEEE Transactions on Software Engineering, 1988
- System Structure Analysis: Clustering with Data BindingsIEEE Transactions on Software Engineering, 1985
- Prototyping Versus Specifying: A Multiproject ExperimentIEEE Transactions on Software Engineering, 1984
- A Successive Approximations Approach to Character WeightingSystematic Zoology, 1969