Software project similarity measurement based on fuzzy C-means

  • Authors:

  • Mohammad Azzeh;Daniel Neagu;Peter Cowling

  • Affiliations:

  • Department of Computing, University of Bradford, Bradford, UK;Department of Computing, University of Bradford, Bradford, UK;Department of Computing, University of Bradford, Bradford, UK

  • Venue:
  • ICSP'08 Proceedings of the Software process, 2008 international conference on Making globally distributed software development a success story
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

A reliable and accurate similarity measurement between two softwareprojects has always been a challenge for analogy-based software cost estimation.Since the effort for a new project is retrieved from similar historicalprojects, it is essentially to use the appropriate similarity measure that findsthose close projects which in turn increases the estimation accuracy. In softwareengineering literature, there is a relatively little research addressed the issue ofhow to find out similarity between two software projects when they are describedby numerical and categorical features. Despite simplicity of exitingsimilarity techniques such as: Euclidean distance, weighted Euclidean distanceand maximum distance, it is hard to deal with categorical features. In this paperwe present two approaches to measure similarity between two software projectsbased on fuzzy C-means clustering and fuzzy logic. The new approaches aresuitable for both numerical and categorical features.