Search

In software engineering, the planning and budgeting stages of a software project are of great importance to all stakeholders, including project managers as well as clients. The estimated costs and scheduling time needed to develop any software project before and/or during startup form the basis of a project's success. The main objective of soft- ware estimation techniques is to determine the actual effort and/or time required for project development. The use of machine learning methods to address the estimation problem has, in general, proven remarkably successful for many engineering problems. In this study, a fully connected neural network (FCNN) model and a metaheuristic, gray wolf optimizer (GWO), called GWO-FC, is proposed to tackle the software development effort estimation (SEE) problem. The GWO is integrated with FCNN to optimize the FCNN parameters in order to enhance the accuracy of the obtained results by improving the FCNN's ability to explore the parameter search field and avoid falling into local optima. The proposed technique was evaluated utilizing various benchmark SEE datasets. Furthermore, various recent algorithms from the literature were employed to verify the GWO-FC performance. In terms of accuracy, comparative outcomes reveal that the GWO-FC performs better than other methods in most datasets and evaluation criteria. Experimental outcomes reveal the strong potential of the GWO-FC method to achieve reliable estimation results. [ABSTRACT FROM AUTHOR]

Correct cost estimates is an essential element to develop economic proposals and being competitive in the market. Companies have to estimate the effort, time and cost before bidding for a project. An inaccurate estimate will lead to money and market share loss. To do that the expected software development size has to be estimated early in the development process. Many software models were presented in the literature to handle this task such as expert judgment, analogy-based estimation, formal estimation models and combination-based estimation models. These models were found to be risky and created many problems in practice related to availability of expertise and inaccurate estimate. Soft Computing techniques were successfully used to solve a diversity of problems in software engineering project cost estimation management. Earlier investigation proved that techniques such as Artificial Neural Networks (ANNs) can solve many problems in the field of software engineering project cost estimation management with promising results. In this paper, we propose two new models for software effort and function point estimation using ANNs. Two types of ANNs will be explored; the Feedforward (FF) and the Radial Basic Function (RBF) neural networks. The Albrecht data set with a number of attributes was used to provide our results. Developed results shows that ANNs models can provide an accurate estimate for both the software effort and number of function points. General Terms: Algorithms, Management, Design, Experimentation, Theory. [ABSTRACT FROM AUTHOR]