期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2014
期号：ICETS
页码：1301
出版社：S&S Publications
摘要：Software Cost Estimation (SCE) can berelated as the process of estimating the most realisticeffort necessary to accomplish a software project. Therapidly improved demand of large-scaled and complexsoftware systems leads managers to settle SCE as oneof the most vital actions that is closely associated withthe success or failure of the whole developmentprocedure. Propose an analytical framework based ona multiple comparisons algorithm in order to rankseveral cost estimation methods, determining thosewhich have important dissimilarity in accuracy, andclustering them in nonoverlapping groups. Toovercome this problem proposed an improved costeffort estimation methods and compared usingappropriate statistical procedures. In this paper wedevelop an intelligent Expert System that supports alltype of software development regardless of their type -either using conventional computer languages orcomponent based visual languages. Classification ismost common method used for finding the mine rulefrom the large database. We also extend our work toC5.0 algorithms applied on customer database forclassification. The proposed framework is applied in alarge-scale setup of comparing 12 prediction modelsover three datasets.
关键词：Software Cost Estimation; C5.0; Expert;System; Scott-Knott; prediction model
