AUTHOREA
Log in Sign Up Browse Preprints

loading page

Evaluating the Accuracy of Different Software Complexity Metrics in Predicting Software Performance.
  • +3
  • H. G. S. Lakshan,
  • D. I. De Silva,
  • M. V. N. Godapitiya,
  • S. P. B. A. Prathyanga,
  • K. L. P. Chamikara,
  • P. S. Wimalagunasekara
H. G. S. Lakshan
Sri Lanka Institute of Information Technology

Corresponding Author:[email protected]

Author Profile
D. I. De Silva
Sri Lanka Institute of Information Technology
Author Profile
M. V. N. Godapitiya
Sri Lanka Institute of Information Technology
Author Profile
S. P. B. A. Prathyanga
Sri Lanka Institute of Information Technology
Author Profile
K. L. P. Chamikara
Sri Lanka Institute of Information Technology
Author Profile
P. S. Wimalagunasekara
Sri Lanka Institute of Information Technology
Author Profile

Abstract

To determine the complexity of software systems, software complexity measures are commonly utilized. However, their accuracy in forecasting software performance has been widely disputed. Accuracy of multiple software complexity measures in forecasting software performance are intended to be examined in this research study, employing numerous performance variables such as execution time, memory utilization, and scalability. Cyclomatic Complexity, Lines of Code, Halstead Complexity measures, and maintainability index are the most used methods to measure the complexity of software system. In this research result indicates how that different complexity matrices are vary in giving different software performance. Also examine what are the limitations of this different complexity matrixes.