Software effort estimation is an important part of the software development
process as it strives to determine the success or failure of the project. The success of the
project is entirely based on the prediction accuracy of the software effort estimation.
There has been a major challenge in agile methodology adoption which is effort
estimation but the conventional means of estimating the effort mainly results in
inaccurate estimates so we need to follow an appropriate model approach. This paper is
reviewed to focus on effort estimation in agile projects which is related to story points
that help to prioritize user stories for faster deployment of the project and to review
different hybrid models that are used to predict the effort. We reviewed the accuracy
parameters based on three popular agile datasets and found that the Deep Belief
Network-Ant Lion Optimizer (DBN-ALO) model works efficiently for all the datasets
and outperforms all the other proposed hybrid models. Different techniques can be used
for minimizing the effort estimation so that the tasks can be done efficiently.
Keywords: Agile Methodology, Ant Lion Optimizer, Deep Belief Network,
Effort Estimation, Story Points.
