No abstract

No abstract

No abstract

No abstract

Estimation of development effort in softwareprojects has been a challenging issue since many years ago.Uncertain behavior of software projects makes the estimatingprocess difficult at early stage of project so that achieving toaccurate estimations seems to be impossible in software projects.Neural Networks(NN) and Analogy Based Estimation (ABE)methods have been widely used in this field because the natureof which is adaptable with dynamic environment of softwareprojects. Since most software project datasets include someirrelevant and inconsistent projects, the quality of neuralnetwork training and also quality of ABE estimations have beensignificant problems in all prior studies. In this paper, toovercome the problem of inconsistence projects, fuzzyclustering has been used for placing the similar projects inseveral clusters. A new framework was proposed to combineABE and NN using C-Means clustering. The results showed thatthe proposed method improved the performance of NN andABE noticeably