A Neuro-Fuzzy Model for Software Cost Estimation

One of the most important tasks of a software development life cycle is to estimate software time and effort. Accuracy in effort estimation is still a great challenge in the software industry. Conventional models like the Constructive Cost Model (COCOMO) are quite helpful but they are unable to manage the unsure nature of the software projects particularly within the early part of the development criterion. There has been considerable work exhausted in this area using neural networks and conjointly some work done using fuzzy logic. These models offer additional exactitude however they " re heavily obsessed on the scale of the training set. The aim of this paper is to develop and evaluate a neuro-fuzzy model to estimate software development effort. Our proposed work is compared with neural network approach and fuzzy logic approach based on the value of MMRE (Mean of Magnitude of Relative Error) .The projected Neuro-fuzzy model was found to have lower MMRE value.

Applied Soft Computing, 2007

Accurate software development cost estimation is important for effective project management such as budgeting, project planning and control. So far, no model has proved to be successful at effectively and consistently predicting software development cost. A novel neuro-fuzzy Constructive Cost Model (COCOMO) is proposed for software cost estimation. This model carries some of the desirable features of a neuro-fuzzy approach, such as learning ability and good interpretability, while maintaining the merits of the COCOMO model. Unlike the standard neural network approach, the proposed model can be interpreted and validated by experts, and has good generalization capability. The model deals effectively with imprecise and uncertain input and enhances the reliability of software cost estimates. In addition, it allows input to have continuous rating values and linguistic values, thus avoiding the problem of similar projects having large different estimated costs. A detailed learning algorithm is also presented in this work. The validation using industry project data shows that the model greatly improves estimation accuracy in comparison with the well-known COCOMO model. #

Software cost estimation is a challenging and onerous task. Estimation by analogy is one of the expedient techniques in software effort estimation field. However, the methodology utilized for the estimation of software effort by analogy is not able to handle the categorical data in an explicit and precise manner. Early software estimation models are based on regression analysis or mathematical derivations. Today's models are based on simulation, neural network, genetic algorithm, soft computing, fuzzy logic modelling etc. This paper aims to utilize a fuzzy logic model to improve the accuracy of software effort estimation. In this approach fuzzy logic is used to fuzzify input parameters of COCOMO II model and the result is defuzzified to get the resultant Effort. Triangular fuzzy numbers are used to represent the linguistic terms in COCOMO II model. The results of this model are compared with COCOMO II and Alaa Sheta Model. The proposed model yields better results in terms of MMRE, PRED(n) and VAF.

2015

Software cost estimation is the process of predicting the effort required to develop a software system. Many estimation models have been proposed over the last 30 years. With the development of software engineering, estimation of project cost and duration has been a very important work. It plays an important role in project bid and project planning. Many papers have been published regarding this topic, which aims at predicting costs of projects to a tolerable degree of accuracy at the early stage. In this paper, several existing fuzzy logic methods for software cost estimation are illustrated and they are compared with the intermediate COCOMO model. Comparing the features of the methods, it could be applied for clustering based on abilities and is also useful for selecting the special method for each project.

2020

Software cost prediction is the technique of accurately evaluating the amount while developing the software. Estimation involves the total time required for the completion of the software, effort required that is measured in terms of person per month (PM), and the total cost to complete the activity. Accuracy and duration are the two desirable criteria in the software estimation process. In software estimation process, there are several inputs that are being fed to the system and these inputs are used for the generation or calculation of the set of outputs. The important work of the software project managers in the present scenario is the computation of cost or effort before the absolute advancement of any particular software. There are several methods applied for software cost estimation but we will focus on the fuzzy logic which is a soft-computing method. We feel that model which is based on fuzzy logic for the software cost estimation should be able to give the uncertain values ...

Managers required administering software development projects with accurate estimates of the resources. So accurate estimates can increase the speed of the effort for developing software projects, and prevent probabilistic failure consequently. Therefore, in the process of software development estimate must be taken in order to reduce cost and time schedule and the existence of probabilistic risks. The implementation of software projects developmental process in large organizations with hundreds of experts and professionals often is a complicated action and needs to be estimated. Estimation cost, time and manpower depends to factors such as the performance of software groups and the complexity of software project. Different models have been proposed for the Software Cost Estimation (SCE) that COCOMO model is the most common model for SCE. This model was used in 1981 by Boehm. This model is a high risk and threat for software projects due to its low accuracy and lack of reliability. Therefore, there is a need to evaluate and estimate SCE using Artificial Intelligence (AI) models. One of the best models of AI for SCE is Artificial Neural Networks (ANNs). ANNs try to assess and evaluate the data with minimal error using instruction and learning techniques. In this paper, the performance of Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), Wavelet Neural Network (WNN), Functional Link Artificial Neural Network (FLANN) and Generalized Regression Neural Network (GRNN) models have been discussed in SCE.

ACM Sigsoft Software Engineering Notes, 2010

Effective Software cost estimation is one of the most challenging and important activities in Software development. The software industry does not estimate projects well. In this paper we have represented size in KLOC as a Fuzzy number. A new model is presented using fuzzy logic to estimate effort required in software development. We use MATLAB for tuning the parameters of famous COCOMO model. The performance of model is evaluated on published software projects data. Comparison of results from our model with existing prevalent models is done.

2000

Abstract Estimation of effort/cost required for development of software products is inherently associated with uncertainty. In this paper, we are concerned with a fuzzy set-based generalization of the COCOMO model (f-COCOMO). The inputs of the standard COCOMO model include an estimation of project size and an evaluation of other parameters. Rather than using a single number, the software size can be regarded as a fuzzy set (fuzzy number) yielding the cost estimate also in form of a fuzzy set.

Quantitative methods have nowadays become very important tools in estimating the effort and cost in software development and support projects. Estimation accuracy is one of the most successful factors for a project's success; hence, the research has not stopped in improving the estimation models. Artificial neural networks (ANNs) are flexible computing frameworks that can be applied to a wide range of forecasting problems with a high degree of accuracy. However, ANNs need a quality training data in order to yield accurate results, data filtering and choosing the right ANN is important for the ANN to gain good results. In this paper, based on the basic concepts of ANNs and the use of Fuzzy regression as data filters a new hybrid method is proposed which yields more accurate results with non-quality data,

Soft Computing-A Fusion of …, 2006

Accurate software estimation such as cost estimation, quality estimation and risk analysis is a major issue in software project management. In this paper, we present a soft computing framework to tackle this challenging problem. We first use a preprocessing neurofuzzy inference system to handle the dependencies among contributing factors and decouple the effects of the contributing factors into individuals. Then we use a neuro-fuzzy bank to calibrate the parameters of contributing factors. In order to extend our framework into fields that lack of an appropriate algorithmic model of their own, we propose a default algorithmic model that can be replaced when a better model is available. One feature of this framework is that the architecture is inherently independent of the choice of algorithmic models or the nature of the estimation problems. By integrating neural networks, fuzzy logic and algorithmic models into one scheme, this framework has learning ability, integration capability of both expert knowledge and project data, good interpretability, and robustness to imprecise and uncertain inputs. Validation using industry project data shows that the framework produces good results when used to predict software cost.