Application of Identical Degree of Set Pair Analysis on Software Refactoring

Set pair analysis posssses many virtues, including specific concept, convenience calculation and it covers comprehensive information as well. In course of software refactoring, traditional code locating method mostly relies on developer's observations and subjective perceptions, which brings with much inconvenience. Therefore, a new software refactoring scheme optimization model was built up based on identical degree of set pair analysis, in virtue of correlation indicators of software metrics, which offered a resolution to code locating in software refactoring. This paper applied variation coefficient method to calculate the weight of indicators, which improves the reliability of scheme evaluation. An experimental case argues that the set pair analysis model can locate the iffy code correctly, and furthermore, it can offer specific refactoring order of each scheme.

[1]  M. Fowler Improving the Design of Existing Code , 2000 .

[2]  Wu Shu-jian Application of principle component analysis and cluster analysis on software refactorings , 2009 .

[3]  Shinji Kusumoto,et al.  Refactoring Effect Estimation Based on Complexity Metrics , 2008, 19th Australian Conference on Software Engineering (aswec 2008).

[4]  Martin Fowler,et al.  Refactoring - Improving the Design of Existing Code , 1999, Addison Wesley object technology series.

[5]  Victor R. Basili,et al.  A Validation of Object-Oriented Design Metrics as Quality Indicators , 1996, IEEE Trans. Software Eng..

[6]  Rushikesh K. Joshi,et al.  Microscopic coupling metrics for refactoring , 2006, Conference on Software Maintenance and Reengineering (CSMR'06).

[7]  Fei Na Object-oriented Software Refactoring , 2005 .

[8]  Letha H. Etzkorn,et al.  Computing software metrics from design documents , 2004, ACM-SE 42.

[9]  Claus Lewerentz,et al.  Metrics Based Refactoring , 2001, CSMR.

[10]  Alexander Chatzigeorgiou,et al.  Identification of Move Method Refactoring Opportunities , 2009, IEEE Transactions on Software Engineering.