TY - JOUR AU - Gora , Rajani Kumari AU - Sinha, Ripu Ranjan PY - 2023/05/17 Y2 - 2024/03/28 TI - A Study of Evaluation Measures for Software Effort Estimation Using Machine Learning JF - International Journal of Intelligent Systems and Applications in Engineering JA - Int J Intell Syst Appl Eng VL - 11 IS - 6s SE - Research Article DO - UR - https://ijisae.org/index.php/IJISAE/article/view/2846 SP - 267 - 275 AB - <p>Software effort estimation is a crucial process which involves predicting how much time and money will be needed to accomplish a software development project. Expert opinion and past data are used in conventional estimating techniques, which may be inefficient and prone to mistakes. Machine learning offers a promising approach to automate this process by learning from past projects and predicting effort estimates for new projects. Using machine learning, this article takes an in-depth look at the practise of software work estimation. In this study, several machine learning models, including support vector machine, KNN, ANN, linear regression, support vector machine, and neural network, are trained and evaluated on a dataset of software projects. This paper also presents some comparative results of the various machine learning algorithms, including multilayer perceptron (MLP), support vector machine (SVM), and linear regression (LR), showing that the MLP model achieves the lowest MMRE value, 13%, while the SVM achieves the highest PRED(25), 87.65%, for software effort estimation.</p> ER -