English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

There is evidence that Software Development Effort Estimation (SDEE) plays a crucial role in managing the software project and controlling its whole lifecycle; an accurate effort estimate allows an effective monitoring and efficient scheduling of tasks and resources. Although extensive research has been carried out on SDEE techniques, no single technique has been shown to be superior to other in all situation. Recently, there has been an increasing amount of literature on predicting software effort using Machine Learning (ML) methods. Among these ML techniques, regression tree-based models have gained a considerable attention due to their generalization ability and understandability. So far, very few studies have investigated the potential of Random Forests (RF) in software effort estimation. In this paper, a RF model is designed and adjusted empirically by varying the values of its key parameters. Prior to the parameters adjustment, we analysed their impact on RF model accuracy which allows an efficient tuning of the model during the training stage. The performance of the RF is then evaluated and compared with that of classical Regression Trees (RT). The evaluation was performed through the 30% hold-out validation method using five datasets: ISBSG R8, Tukutuku, COCOMO, Desharnais and Albrecht. To identify the most accurate technique, we employed three widely known accuracy measures: Pred(0.25), MMRE and MdMRE. The results obtained show that the adjusted random forest outperforms the regression trees model on all evaluation criteria. Moreover, the proposed model performs better than some recent techniques reported in the literature for software effort estimation.

Software Development Effort Estimation Using Random Forests: An Empirical Study and Evaluation