文章基本信息

标题：Comparing Software Cost Estimation Using Scott-Knott Test
本地全文：下载
作者：M.Sujitha ; N.Sivakumar
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：2015
期号：MULTICON
页码：114
出版社：S&S Publications
摘要：Software cost Estimation is the approximate judgement of the costs for a project. It is to predict the mostrealistic effort needed to complete a software project. Accurate effort prediction is done by comparing the traineddataset through multiple algorithms. In this paper, the proposed cost estimation model identifies the differences inaccuracy of data set, and also clusters it into non-overlapping groups. Specific technologies are not bound in modernsoftware development methodologies. To overcome this problem a system is proposed to improve cost effort estimationmethods and then compared using appropriate statistical procedures for ensuring the appropriate results. The ScottKnott test is to perform multiple comparisons without ambiguity. The proposed modification related to the partitioningand means grouping to obtain results without ambiguity among models. In the proposed methodology, models that didnot participate in the initial group are joined for a new analysis, which allows for a better group distribution.
关键词：software cost estimation; software metrics; software effort estimation; statistical methods