Empirical evaluation of the active learning strategies on software defects prediction

We present a Comprehensive Fault Prediction Model (FPM) for Software Reliability in this article, which can estimate the greatest quantity of faults in a software. The proposed model implemented on a “Non-Homogeneous Poisson Process Model (NHPPM)” and includes fault reliant identification, rate of failure, and the software defect present after release. We looked at programmers' abilities, software performance, and flawless debugging as deciding factors for FPM. The Software Engineering Team Assessment and Prediction (SETAP) dataset is used for analysis of proposed FPM. The selected dataset is composed of sequential value which are linearly arranged over a given time duration. The attributes are analyzed to establish software reliability prediction model and comparison of proposed model is carried out with similar algorithms. The proposed FPM is executed in “Jira” and is compared with the present FPMs proposed in the literature.Results demonstrate comparatively less cumulative faults, and reduced residual errors which depicts high prediction accuracy and improved software reliability.

show abstract

Empirical evaluation of the active learning strategies on software defects prediction

Cited by 3 publications

References 13 publications

Software Defect Prediction Using ROS-KPCA Stacked Generalization Model

Software Defect Prediction Using ROS-KPCA Stacked Generalization Model

Data-efficient software defect prediction: A comparative analysis of active learning-enhanced models and voting ensembles

A Comprehensive Fault Prediction Model for Improving Software Reliability

Contact Info

Product

Resources

About