अमूर्त

Intelligent Software Effort Estimation through a Multiple Comparisons Algorithm

R.Manimegalai, J.Selvakumar, M.Rajaram

Software Cost Estimation (SCE) can be related as the process of estimating the most realistic effort necessary to accomplish a software project. The rapidly improved demand of large-scaled and complex software systems leads managers to settle SCE as one of the most vital actions that is closely associated with the success or failure of the whole development procedure. Propose an analytical framework based on a multiple comparisons algorithm in order to rank several cost estimation methods, determining those which have important dissimilarity in accuracy, and clustering them in nonoverlapping groups. To overcome this problem proposed an improved cost effort estimation methods and compared using appropriate statistical procedures. In this paper we develop an intelligent Expert System that supports all type of software development regardless of their type - either using conventional computer languages or component based visual languages. Classification is most common method used for finding the mine rule from the large database. We also extend our work to C5.0 algorithms applied on customer database for classification. The proposed framework is applied in a large-scale setup of comparing 12 prediction models over three datasets.