Efficient estimation of effort using machine-learning technique for software cost

Abstract

Several useful models have been developed by the software engineering community to elucidate the periodic growth of life cycle and calculate the effort of cost estimation in a precise manner. One of the commonly used machine learning techniques is the analogy method that cannot handle the categorical variables efficiently. In general, project attributes of cost estimation are often measured in terms of linguistic values. These imprecise values leads to analogous while explaining the process. The proposed fuzzy analogy method is a new approach based on reasoning by analogy using fuzzy logic for handling both numerical and categorical variables where the uncertainty and imprecision solution is also identified by the behavior of linguistic values utilized in the software projects. The performance of this method validates the results based on historical NASA dataset. The outcome of fuzzy analogy method is analyzed which indicates its improvement over the existing fuzzy logic methods.