Software Project Estimation Using Machine Learning
This method enables us to successfully combine Machine Learning with the Analogous Estimation methodology, thereby boosting the precision of software project estimates for more dependable outcomes.
Abstract
The major purpose of the project is to investigate a range of machine learning techniques in order to improve the accuracy of projecting costs and efforts for software development projects. Numerous strategies in software effort estimation have been investigated in order to develop models with the best accuracy. Our primary goal is to determine the important variables and factors that influence the costs and efforts necessary for software development. To solve the accuracy issues, we're building an Extreme Learning Machine (ELM) model that uses the Analogous Estimation technique. The performance of the model will then be compared to that of existing models such as the Constructive Cost Model (COCOMO), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM). This method enables us to successfully combine Machine Learning with the Analogous Estimation methodology, thereby boosting the precision of software project estimates for more dependable outcomes.