원문정보
초록
영어
Trending strategy for better management of system complexity is following object-oriented approach in development of software. Object-oriented (OO) approach increases the chances of upgrading quality. Few researches have been in direction of OO software development and metrics utility. Also, evidence shows for inadequate of empirically linking the OO methodology and project outcomes. Need of hour is advancement in research for better understanding the parameters influencing software quality and other project effects such as productivity in OO software development. This has been accepted as one of the prevalent challenges in OO methodology and is trying to resolve these issues. This paper discusses the role of CK metric suite in elucidating object-oriented software bugs at a class level. The analyses deals with projects following OO approach during development highlighting the correlation of the metrics in CK suite with the bugs present. The motivational drive is to explicitly accounting that larger classes will have more number of bugs and we tried to develop model based on certain proposed hypotheses. Our approach is to study the effect of complexity that arises due to interaction among various CK metrics suite and bugs.
목차
1. Introduction
2. Prior Literature
2.1. Metrics for OO Design Complexity
2.2. Background on CK Metric
2.3. Exploration using Metrics
3. Model and Hypotheses
4. Data Collection
5. Empirical Model and Diagnostics
5.1. Model Specification
5.2. Regression Diagnostics
6. Discussion of Results
6.1. Impact of Class Size
6.2. Impact of Weighted Methods per Class (WMC)
6.3. Impact of Depth Inheritance Tree per Class (DIT)
6.4. Impact of Coupling between Objects per Class (CBO)
6.5. Impact of Coupling between Objects (CBO) and Depth of Inheritance (DIT)
6.6. Impact of Coupling between Objects (CBO) and Weighted Methods per Class (WMC)
7. Conclusions
Appendix
References