Multi-variate principal component analysis of software maintenance effort drivers

  • Authors:

  • Ruchi Shukla;A. K. Misra

  • Affiliations:

  • Computer Science and Engineering Department, Motilal Nehru National Institute of Technology, Allahabad, India;Computer Science and Engineering Department, Motilal Nehru National Institute of Technology, Allahabad, India

  • Venue:
  • ICCSA'10 Proceedings of the 2010 international conference on Computational Science and Its Applications - Volume Part II
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The global IT industry has already attained maturity and the number of software systems entering into the maintenance stage is steadily increasing. Further, the industry is also facing a definite shift from traditional environment of legacy softwares to newer softwares. Software maintenance (SM) effort estimation has become one of the most challenging tasks owing to the wide variety of projects and dynamics of the SM environment. Thus the real challenge lies in understanding the role of a large number of SM effort drivers. This work presents a multi-variate analysis of the effect of various drivers on maintenance effort using the Principal Component Analysis (PCA) approach. PCA allows reduction of data into a smaller number of components and its alternate interpretation by analysing the data covariance. The analysis is based on an available real life dataset of 14 drivers influencing the effort of 36 SM projects, as estimated by 6 experts.