Abstract
Empirical Application of Simulated Annealing Using Object-Oriented Metrics to Increase the Accuracy in Software Effort Estimation
Highlights
Software Effort Estimation [1] is the approximate calculation about the work to be performed in order to develop a software project
Since in our previous discussion we described effort estimation model, the simulated annealing (SA) based estimation relations were derived as a function dependent of line of codes (LOC) and Effort Adjustment Factor (EAF) for COCOMO81 dataset
On comparing SA with Artificial Neural Network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS) techniques for Software Development Effort estimation model development, it is seen that Neuro-fuzzy method ANFIS presented in this thesis shows a good potential to model complex, nonlinear and multivariate problems
Summary
Software Effort Estimation [1] is the approximate calculation about the work to be performed in order to develop a software project. Lack of knowledge about converting the human expertise into knowledge base of FIS and secondly, in order to reduce the difference between the observed values and predicted values, there is need for learning algorithm for tuning the membership function [8][9] These two problems greatly restrict the application domains of FIS. Neural Network modelling does not rely on human expertise Instead, it employs a learning procedure and a given training data set to solve a set of parameters ( i.e. weights) such that the required functional behaviour is achieved. The round hubs speak to hubs that are altered while the square hubs are hubs that have parameters to be learnt
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have