Software effort estimation accuracy prediction of machine learning techniques: A systematic performance evaluation

Software effort estimation accuracy is a key factor in effective planning, controlling, and delivering successful software project within budget schedule. The overestimation underestimation both are the challenges for future development, henceforth there continuous need estimation. researchers practitioners striving to identify which machine learning technique gives more accurate results based on evaluation measures, datasets other relevant attributes. authors of related research generally not aware previously published techniques. main aim this study assist know yields promising prediction development. In article, performance ensemble solo techniques investigated publicly non-publicly domain two most commonly used metrics. We systematic literature review methodology proposed by Kitchenham Charters. This includes searching papers, applying quality assessment (QA) criteria, extracting data, drawing results. have evaluated state-of-the-art 35 selected studies (17 ensemble, 18 solo) using mean magnitude relative error PRED (25) as set reliable metrics among report questions stated study. found that frequently implemented construction (EEE) revealed EEE usually yield than