Minimizing Analogy Errors with the Help of Fuzzy

摘要

The current study is about minimizing errors in software cost estimation when analogy technique is used for the estimation. Project and product based companies face lot of problems during cost estimation because of requirement changes from time to time which in-turn changes the utility cost. Overestimation and underestimation lead to complete collapse and failure of the project at earlier stages itself. Analogy based estimation is considered to be the best methodology in algorithmic modeling. Analogy is to create accurate estimates for the proposed project by comparing the proposed project to similar projects from the past. While doing a comparison in analogy we face lot of problems at runtime. To overcome analogy errors such as observational error, trapezoidal membership function is used. When an estimator is new to analogy, the person is prone to become suspicious about choosing the right set of projects to derive a comparison. In this paper, we analyze how S-Membership function could be used to deal with such scenarios.