根据序列变化率预测软件阶段成本.pdfVIP

第 32 卷第 7 期 计算机学报 Vo1. 32 No. 7 2009 年 7 月 CHINESE JOURNAL OF COMPUTERS July 2009 根据序列变化率预测软件阶段成本 王勇宋擒豹沈钧毅 〈西安交通大学计算机科学与技术系 西安 710049) 摘 要 针对软件阶段成本因少数据、不确定性使得用现有方法〈如回归)难以预测的问题，文中提出一种新颖的 预测方法，该方法从项目已完成阶段的成本序列中，通过变换得到反映序列变化快慢的变化率，并用机器学习方 法从历史项目中学习得到变化率阙值，然后用不同的灰色模型进行预测.在 10 个现实世界软件工程数据集上的实 验结果表明，该方法平均预测误差比线性回归方法低 20%-80% ，显示出较大的潜力. 关键词 项目成本预测;软件成本;阶段成本 z 灰色模型 中圄法分类号 TP311 DOI 号: 10. 3724jSP.J. 1016.2009. 01346 Predicting Software Stage Effort with Sequence Changing Ratio WANG Yong SONG Qin-Bao SHEN Jun-Yi (De户artment 01 Computer Science and Technology ， 刀 an J iaotong University. 沽 an 710049) Abstract Software stage effort has the features of data starvation and uncertainty. It is difficult to use the current methods (e. g. regression) to make predictions. This paper proposes a novel prediction method, which gets the effort sequence changing feature 一 changing ratio from the completed stage effort sequences , and gets the changing ratio threshold from historical projects by machine learning methods, then uses grey models to make predictions. The experimental re? sults on 10 real world software engineering datasets show that , compared with linear regression method , the prediction accuracy of the proposed method has been improved by 20 %~ 80 %. This is very encouraging and indicates that the method has considerable potential. Keywords project cost prediction; software cost; stage cost; grey model 士一同 现可能出现的成本进度方面的问题，提前采取有效 1 百|