One of the most important objectives of software engineering community hasbeen the increase of useful models that beneficially explain the development oflife cycle and precisely calculate the effort of software cost estimation. Inanalogy concept, there is deficiency in handling the datasets containingcategorical variables though there are innumerable methods to estimate thecost. Due to the nature of software engineering domain, generally projectattributes are often measured in terms of linguistic values such as very low,low, high and very high. The imprecise nature of such value represents theuncertainty and vagueness in their elucidation. However, there is no efficientmethod that can directly deal with the categorical variables and tolerate suchimprecision and uncertainty without taking the classical intervals and numericvalue approaches. In this paper, a new approach for optimization based on fuzzylogic, linguistic quantifiers and analogy based reasoning is proposed toimprove the performance of the effort in software project when they aredescribed in either numerical or categorical data. The performance of thisproposed method exemplifies a pragmatic validation based on the historical NASAdataset. The results were analyzed using the prediction criterion and indicatesthat the proposed method can produce more explainable results than othermachine learning methods.
展开▼
