Predicting Project Effort Intelligently in early Stages by Applying Genetic Algorithms with Neural Networks

2014 ◽  
Vol 513-517 ◽  
pp. 2035-2040 ◽  
Author(s):  
Zhen You Li
Keyword(s):  
Neural Networks ◽  
Genetic Algorithms ◽  
Cost Estimation ◽  
Back Propagation ◽  
Development Project ◽  
Software Cost ◽  
Reduced Models ◽  
Software Development Project ◽  
Early Stages ◽  
Global Optimal

In the early stages of a software development project, estimating the amount of effort is one of the most important project management concerns. This study has successfully produced global optimal reduced models intelligently predicting software cost estimation by employing neural networks with back-propagation learning algorithms combined with genetic algorithms (GA-NN) to determine the most significant explanatory variables among the 16 COCOMO cost drivers. The performance of the full model of GA-NN is much superior to that of the COCOMO, whilst the predicting performance of its global optimal reduced model is also comparable to that of the COCOMO in terms of MMRE and PRED (25). The optimal reduced models and their found significant factors can offer powerful supports for the project managers to make right decisions in the early stages of the projects.

scholarly journals Risk Analysis in Software Cost Estimation: A Simulation-Based Approach

2021 ◽  
Vol 12 (6) ◽  
pp. 2176-2183
Author(s):  
Vikram Singh, Varun Malik, Ruchi Mittal
Keyword(s):  
Risk Analysis ◽  
Cost Estimation ◽  
Development Project ◽  
Project Planning ◽  
Software Cost Estimation ◽  
Present Communication ◽  
Software Cost ◽  
Software Development Project ◽  
Simulation Based ◽  
The Way

Risk analysis and cost estimation are two important aspects of project planning that can either make the way or break the way to a project’s success. At the same, both these tasks are difficult and painstaking, but whether someone likes it and not, the project’s success depends heavily on them. As documented by Fredric Brooks Junior in his legendry book “The Mythical Man-Month,” planning, scheduling, and estimation have been central to software engineering since its early days in the 1970s. Present communication presents a simulation-based approach to estimate the costing schedule of a software development project. The results show that simulation is expedient as well as efficient in terms of time, effort, and cost requirement and provides pragmatic results.

Kernel Fuzzy Clustering with Output Layer Self-Connection Recurrent Neural Networks for Software Cost Estimation

2019 ◽  
Vol 29 (06) ◽  
pp. 2050091
Author(s):  
V. Resmi ◽  
S. Vijayalakshmi
Keyword(s):  
Neural Networks ◽  
Fuzzy Clustering ◽  
Recurrent Neural Networks ◽  
Cost Estimation ◽  
Clustering Algorithm ◽  
Estimation Problem ◽  
Software Cost Estimation ◽  
Software Cost ◽  
Output Layer ◽  
Fuzzy C Means Clustering

In the current world, the software cost estimation problem has been resolved using various newly developed methods. Significantly, the software cost estimation problems can be dealt with effectively with the recently grown recurrent neural network (RNN) than the other newly developed methods. In this paper, an improved approach is proposed to software cost estimation using Output layer self-connection recurrent neural networks (OLSRNN) with kernel fuzzy c-means clustering (KFCM). The proposed OLSRNN method follows the basics of traditional RNN models for integrating self-connections to the output layer; thereby, the output temporal dependencies are better captured. Also, the performance of neural networks is improved using the kernel fuzzy clustering algorithm to enhance software estimation results. Ultimately, five publicly available software cost estimation datasets are adapted to verify the efficacy of the proposed KFCM-OLSRNN method using the validation metrics such as MdMRE, PRED (0.25) and MMRE. The experimental results proved the efficiency of the proposed method for solving the software cost estimation problem.

scholarly journals Sign Language Recognition using Hybrid Neural Networks

2019 ◽  
Vol 9 (2) ◽  
pp. 1092-1098
Keyword(s):  
Neural Networks ◽  
Genetic Algorithms ◽  
Artificial Neural Networks ◽  
Sign Language ◽  
Back Propagation ◽  
Optimization Techniques ◽  
Language Recognition ◽  
Sign Language Recognition ◽  
Sign Languages ◽  
Artificial Neural

Language has a prime role in communication between persons, in learning, in distribution of concepts and in preserving public contacts. The hearing-impaired have to challenge communication obstacles in a mostly hearing-capable culture. There are hundreds Sign Languages that are used all around the world today .The Sign Languages are established depending on the country and area of the deaf public. The aim of sign language recognition is to offer an effectual and correct tool to transcribe hand gesture into text. It can play a vital role in the communiqué between deaf and hearing society. Sign language recognition (SLR), as one of the significant research fields of human–computer interaction (HCI), has produced more and more interest in HCI society. Since, artificial neural networks are best suited for automated pattern recognition problems; they are used as a classification tool for this research. Back propagation is the most important algorithm for training neural networks. But, it easily gets trapped in local minima leading to inaccurate solutions. Therefore, some global search and optimization techniques were required to hybridize with artificial neural networks. One such technique is Genetic algorithms that imitate the principle of natural evolution. So, in this article, a hybrid intelligent system is proposed for sign language recognition in which artificial neural networks are merged with genetic algorithms. Results show that proposed hybrid model outperformed the existing back propagation based system.

Sign in / Sign up

Export Citation Format

Share Document