Software Effort Estimation Using Synthetic Minority Over-Sampling Technique for Regression (SMOTER)
Misha Jawa, Shweta Meena · 2022 3rd International Conference for Emerging Technology (INCET) · 2022
Estimating the effort of a software project is one of the initial steps in software project development. Effort estimation is a practice of estimating how much effort will be necessary to develop or maintain a software project. It is important to calculate the effort at early stage in order to deliver the project on time without compromising the quality of project. The main goal of our research is to study the impact of Synthetic Minority Over-Sampling Technique for Regression (SMOTER) to predict effort estimation by using machine learning algorithms. The different machine learning algorithms that we have used are decision tree regressor, random forest, linear regression, lasso regression and ridge regression. In our study we used three datasets i.e., China, Maxwell and COCOMO81. The two-performance metrics Mean Magnitude Relative Error (MMRE) and PRED(25) were used to compare the results of each model using SMOTER and without using SMOTER. We have seen a significant decrease in error of each model after applying SMOTER.