Software Effort Estimation for COCOMO-II Projects Using Artificial Neural Network
- July 2, 2020
- Posted by: RSIS
- Categories: Computer Science and Engineering, IJRSI
International Journal of Research and Scientific Innovation (IJRSI) | Volume VII, Issue VI, June 2020 | ISSN 2321–2705
Software Effort Estimation for COCOMO-II Projects Using Artificial Neural Network
Kiran Kumar T.M1*, Yashvanth Kumar K.P2
1Assistant Professor, 2Project Student, Dept of MCA, Siddaganga Institute of Technology, Tumkur, India
Abstract: Software failures are mainly caused by defective projects management practices, including estimates of effort. Constant changing outlines of requirements and the technology software development make estimating efforts more complicated. Several methods are available to Estimate the effort of the soft computing-based method. The development effort needed for a project should be measured by software. It is important to estimate the construction effort required before any project is initially initiated. It is one of the greatest and most demanding tasks ever. The software cost estimate deals with a lot of uncertainty between all neural computing methods. In this paper we have used the historical COCOMO II data set projects using the artificial neural network technique to predict the effort estimation. We have used the mat lab tool for estimation. The experiment outputs suggest that the suggested model can provide better results and accurately forecast the software development effort.
Keywords: Artificial Neural Networks; Back Propagation; COCOMO II; Effort Estimation; MRE.
I. INTRODUCTION
The main objective of any software industry is to develop the product on time within the budget which is relevant in practice. Effort evaluation is a mechanism by which development time and costs for software or product development can be estimated. Cost and time estimates for a variety of purposes, precisely is important. More than an estimate could hurt a company’s financial loss.
Effort estimates made during the initial stages of project development can also be useful for project managers. However, at early stages, there is very little detail. Some algorithmic and non-algorithmic methods for estimation of effort already exist. Since the requirements are volatile and the complex environments the predictions of effort is still challenging task for any project managers in the software industry.
Correctness in estimating the software development effort required plays a critical role in the success of software project management a correct resource amount should be provided for a project. The calculated effort will be approximately equivalent to the true effort for a successful software evaluation method.
