Using Tabu Search to Estimate Software Development Effort

  • Authors:

  • Filomena Ferrucci;Carmine Gravino;Rocco Oliveto;Federica Sarro

  • Affiliations:

  • Dipartimento di Matematica e Informatica, University of Salerno, Fisciano, Italy I-84084;Dipartimento di Matematica e Informatica, University of Salerno, Fisciano, Italy I-84084;Dipartimento di Matematica e Informatica, University of Salerno, Fisciano, Italy I-84084;Dipartimento di Matematica e Informatica, University of Salerno, Fisciano, Italy I-84084

  • Venue:
  • IWSM '09 /Mensura '09 Proceedings of the International Conferences on Software Process and Product Measurement
  • Year:
  • 2009

Quantified Score

Hi-index 0.01

Visualization

Abstract

The use of optimization techniques has been recently proposed to build models for software development effort estimation. In particular, some studies have been carried out using search-based techniques, such as genetic programming, and the results reported seem to be promising. At the best of our knowledge nobody has analyzed the effectiveness of Tabu search for development effort estimation. Tabu search is a meta-heuristic approach successful used to address several optimization problems. In this paper we report on an empirical analysis carried out exploiting Tabu Search on a publicity available dataset, i.e., Desharnais dataset. The achieved results show that Tabu Search provides estimates comparable with those achieved with some widely used estimation techniques.