Research Article
Proposing a Novel Artificial Neural Network Prediction Model to Improve the Precision of Software Effort Estimation
@INPROCEEDINGS{10.1007/978-3-642-32615-8_33, author={Iman Attarzadeh and Siew Ow}, title={Proposing a Novel Artificial Neural Network Prediction Model to Improve the Precision of Software Effort Estimation}, proceedings={Bio-Inspired Models of Network, Information, and Computing Systems. 5th International ICST Conference, BIONETICS 2010, Boston, USA, December 1-3, 2010, Revised Selected Papers}, proceedings_a={BIONETICS}, year={2012}, month={10}, keywords={Software engineering software project management software cost estimation models COCOMO model soft computing techniques artificial neural networks}, doi={10.1007/978-3-642-32615-8_33} }- Iman Attarzadeh
Siew Ow
Year: 2012
Proposing a Novel Artificial Neural Network Prediction Model to Improve the Precision of Software Effort Estimation
BIONETICS
Springer
DOI: 10.1007/978-3-642-32615-8_33
Abstract
Nowadays, software companies have to mange different software development processes based on different time, cost, and number of staff sequentially, which is a very complex task and supports project planning and tracking. Software time, cost and manpower estimation for separate projects is one of the critical and crucial tasks for project managers. Accurate software estimation at an early stage of project planning is counted as a great challenge in software project management, in the last decade, as it allows considering project financial, controlling, and strategic planning. Software effort estimation refers to the estimations of the likely amount of cost, schedule, and manpower required to develop software. This paper proposes a novel artificial neural network prediction model incorporating Constructive Cost Model (COCOMO). The new model uses the desirable features of artificial neural networks such as learning ability, while maintaining the merits of the COCOMO model. This model deals efficiently with uncertainty of software metrics to improve the accuracy of estimates. The experimental results show that using the proposed model improves the accuracy of the estimates, 8.36% improvement, when the obtained result compared to the COCOMO model.
