PCA Analysis of Few Parameters Role in Software Development

Abstract

Principal component analysis is one of the dimension reduction methods with the goal of using the correlation structure among the predictor variables. Qualitative/quantitative measurement of software quality related aspects in all stages of software development are desirable [9,10,11,12].Any measurement using any element in the software metrics is helpful for analysis in the set of software quality metrics. In this paper seventeen software metric variables [4], are considered. Four cases are carried out using principle component analysis. First analysis is with size as predominant factor. Second analysis is with effort as predominant factor. Third analysis is with duration as predominant factor. Finally all the three association taken together used for analysis of quality performance. The analysis of variables is to identify the dimension that are latent [7,8].