Version 1
: Received: 6 March 2024 / Approved: 7 March 2024 / Online: 8 March 2024 (03:47:24 CET)
How to cite:
Oshaibi M, F.; AlKhanafseh, M.; Surakhi, O. Software Effort Estimation using Ensemble Learning. Preprints2024, 2024030437. https://doi.org/10.20944/preprints202403.0437.v1
Oshaibi M, F.; AlKhanafseh, M.; Surakhi, O. Software Effort Estimation using Ensemble Learning. Preprints 2024, 2024030437. https://doi.org/10.20944/preprints202403.0437.v1
Oshaibi M, F.; AlKhanafseh, M.; Surakhi, O. Software Effort Estimation using Ensemble Learning. Preprints2024, 2024030437. https://doi.org/10.20944/preprints202403.0437.v1
APA Style
Oshaibi M, F., AlKhanafseh, M., & Surakhi, O. (2024). Software Effort Estimation using Ensemble Learning. Preprints. https://doi.org/10.20944/preprints202403.0437.v1
Chicago/Turabian Style
Oshaibi M, F., Mohammad AlKhanafseh and Ola Surakhi. 2024 "Software Effort Estimation using Ensemble Learning" Preprints. https://doi.org/10.20944/preprints202403.0437.v1
Abstract
Software project estimating (SPE) is a powerful but necessary step in the production of projects for software. Software project estimating includes several components, estimating software time, software resources, software costs, and software effort estimation (SEE). SEE seeks to estimate the number of "hours-person" and "months-person" of work required for developing or managing a software program. During the initial phase of the development of software, future efforts are impossible to predict. To estimate effort, many deep learning (DL) and machine learning (ML) models were previously created. For estimate, single-model techniques and ensemble methods were investigated in this research. Ensemble methods involve a combination of multiple separate models. Bagging, boosting, and voting were the ensemble methods looked into for estimation. Datasets considered for estimation were COCOMO81, COCOMONasa I, COCOMONasa II, Albrecht, Maxwell, China, Desharnais, Desharnais_1_1, Kitchenham, Boehm, and Belady. The evaluation criteria used MAE, RAE, RMSE, RRSE, and CC. The result from previous studies is that the total size of the set of data utilized is small in predicting the effort, as the largest size of the set of data used is the China dataset containing only 499 projects. So, in this study, the dataset has been merged to increase the total size of the set of data and increase the accuracy of prediction. The results indicated that the suggested merged dataset increased effort estimation accuracy.
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
