AN IMPROVED TECHNIQUE FOR SOFTWARE COST ESTIMATIONS IN AGILE SOFTWARE DEVELOPMENT USING SOFT COMPUTING TECHNIQUES

Main Article Content

Quazi Bushra, et. al.

Abstract

The management & estimation of agile projects is stimulating works for many software companies for their high failure rates. To develop successful software projects. Proper prediction of projects overall effort & cost evaluation is a very important task. The numbers of development models over the last few decades have evolved through software projects. Hence, to complete an exact estimation of exertion & taken a toll for diverse program ventures which is based on distinctive improvement models are having innovative & new steps of software development is a significant task which is to be done. Software companies have adopted different various development models which are based on the organization and requirement of project. In this paper we proposed a COCOMO (Constructive Cost Model) for cost estimation of better software projects. Profit or loss estimation forecast to new project is carried out with the help of historical data of company. In the machine learning to predict forecast using historic data Naïve Bayes algorithm plays vital role and provides great accuracy. To check the behavior of the proposed system here we have used the SEERA dataset. According to the result our proposed system gives the profit and loss forecast prediction with the accuracy of 86.59% and 24.80% respectively. And the overall effort calculation accuracy is higher, 95.06% in the contrast to the SVM, 93.45%.

Article Details

Section
Articles