An Evolutionary Computation Approach for Project Selection in Analogy based Software Effort Estimation

Objectives: Software effort estimation is a critical task in the software development process due to the intangible nature of software. A new model for software effort estimation using Differential Evolution Algorithm called DEAPS is proposed in this paper. Method: In this methodology, the complete set of historical project base is reduced to a set of similar projects using Euclidean distance metric. Then Differential Evolution Algorithm which is an Evolutionary Computation method is used for optimization and the most relevant project is retrieved. The proposed method is validated on Desharnais dataset. Findings: DE has a very effective mutation process which improves the ability of exploration. So we got promising results which indicate that the use of this model could significantly improve the efficiency of Analogy based Software Effort Estimation. The metrics used are MMRE, MdMRE and pred (25%). The results are compared with previous findings and the results clearly show that the proposed method is better than the existing methods. Application: This methodology can be used to minimize the errors in the software estimation so that financial loss and delay in the completion of project may be avoided.