RETRACTED ARTICLE: Ensemble learning with recursive feature elimination integrated software effort estimation: a novel approach
| Autor: | K. Eswara Rao, G. Appa Rao |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Estimation
Computer science business.industry COCOMO Cognitive Neuroscience media_common.quotation_subject Integrated software 020206 networking & telecommunications 02 engineering and technology Machine learning computer.software_genre Ensemble learning Mathematics (miscellaneous) Software Resource (project management) Artificial Intelligence Feature (computer vision) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Quality (business) Computer Vision and Pattern Recognition Artificial intelligence business computer media_common |
| Zdroj: | Evolutionary Intelligence. 14:151-162 |
| ISSN: | 1864-5917 1864-5909 |
| DOI: | 10.1007/s12065-020-00360-5 |
| Popis: | To develop software, estimating actual effort is important for any organization as there is no chance of getting either overestimation or underestimation. Due to the overestimation of effort, there may be an immediate need to compromise with the quality and testing. Similarly, underestimation may lead to allocating more resource. Compared to some of the early developed estimation techniques, machine learning based approaches are keen to estimate the effort more accurately due to their dynamic adaptivity with any type of data. With the rapid development of software products, many methods fail to satisfy the objective of development in an effective way. In this paper, a novel model based on ensemble learning and recursive feature elimination based method has been proposed to estimate the effort. With the feature ranking and selection method, the proposed method is able to estimate the efforts with the parameters like size and cost. Simulation results are encouraging with the proposed method with COCOMO II dataset. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink