Detection of Aberrant Data Points for an effective Effort Estimation using an Enhanced Algorithm with Adaptive Features
Software cost estimation plays a critical role to predict the effort and evaluate the feasibility of the project based on the costs involved. It is invariably essential that the technique used to estimate the cost should produce accurate results for decision making. Numerous methods have been proposed for the effort estimation, which fall into one of the three broad categories viz., expert judgment, algorithmic models and machine learning (Mendes et al., 2003). This study presents an enhanced algorithm that predicts the software cost using the Analogy-X method by identifying the most appropriate and stable set of project sets which aid in the prediction of the effort involved.