Effort Estimation in Agile Software Development Using Deep Learning Model
Khalid Alsubhi · International Journal of Advances in Computer Science and Technology · 2019
Software effort estimation is a major activity in the process of software development that permits managers and software engineers to accurately estimate the schedule and budget.Although traditional effort estimation approaches are used to estimate effort for agile software projects but they mostly result in inaccurate estimates.According to standard surveys, 30% to 40% of software projects failed because of inaccuracy of software effort predictions.To improve the accuracy and efficiency of effort estimation, an agile software development process is introduced and substitutes the traditional methods in the software industry.One approach of calculating effort of agile projects is the Story Point Approach.Software industries have adapted the Agile model where effort is estimated using Story Points.The goal of this study is to provide a detailed overview of the effort estimation techniques in Agile Software Development based on Story Points.In this paper, we propose a prediction model for estimating story points based on long short-term memory and recurrent network.This model has been applied to a dataset composed of issue reports mined from JIRA repositories.The proposed work uses three different performance metrics i.e. mean magnitude of relative error, mean square error, and prediction to evaluate the performance of the model.The results of the proposed models are compared to the existing models in the literature.