标题:
基于主成分分析的GRNN在软件开发成本预测的应用
Application of General Regression Neural Network in Software Development Cost Estimation Based on Principle Component Analysis
作者:
陈妤嘉, 罗荣华
关键字:
软件开发成本, 主成分分析, 广义回归类神经网络
Software Development Costs; Principle Component Analysis; General Regression Neural Network
期刊名称:
《Software Engineering and Applications》, Vol.2 No.3, 2013-06-24
摘要:
软件项目管理的质量与开发成本，决定了软件项目的成功与否，时间、质量与成本项目的考虑，成为了影响软件开发成本的关键性要素，项目经理在面对软件开发项目时，即需要预估合理的软件开发成本，现行软件产业最普遍使用的软件成本预估方法，多以项目经理的经验为基础，参考过去开发过的软件项目数据作为主观的项目预估值，本研究提出一个能运行于软件项目开发成本预测模型之研究，运用主成分分析与广义回归类神经网络结合的一种新的预测方法。The quality of project management and project cost are important factors affecting the success of software projects. These critical factors for software development contain time, quality and cost. Nowadays, the most popular method for software development cost estimation is judged by the project manager’s experience. The project manager needs to estimate a reasonable software development cost according to the previous relevant data information of project while facing the problem. Therefore, in this research we propose a new cost estimation model based on the Principle Component Analysis (PCA) and General Regression Neural Network (GRNN) for software development project.