上海交通大学学报(自然版) ›› 2012, Vol. 46 ›› Issue (11): 1736-1740.

• 自动化技术、计算机技术 • 上一篇    下一篇

软件成本数据的相似性度量

刘海涛a,魏汝祥b,蒋国萍a   

  1. (海军工程大学 a.装备经济管理系; b.理学院, 武汉 430033)  
  • 收稿日期:2011-12-28 出版日期:2012-11-30 发布日期:2012-11-30
  • 基金资助:

    海军工程大学自然科学基金引导项目(HGDYDJJ11011)

Similarity Measurement of Software Cost Data

 LIU  Hai-Tao-a, WEI  Ru-Xiang-b, JIANG  Guo-Ping-a   

  1. (a. Department of Equipment Economics and Management; b. College of Science, Naval University of Engineering, Wuhan 430033, China)
  • Received:2011-12-28 Online:2012-11-30 Published:2012-11-30

摘要: 软件成本数据常常表现为高维混合属性数据,传统的相似性度量已不再适用.文中通过建立软件成本数据的高维模糊C均值(FCM)聚类算法对数据相似性进行度量.首先,定义由序数属性到数值属性的初始映射;然后,通过建立改进的迭代高维FCM聚类算法对序数数值映射进行修正,优化聚类效果;最后,利用得到的模糊划分矩阵对软件成本数据的相似性进行度量.实验结果表明,通过对聚类效果进行优化,文中定义的相似性度量能够提高软件成本估算精度.

关键词: 软件工程, 相似性度量, 高维度, 序数属性

Abstract: For the characters of high dimension and mixed attributes of software cost data, traditional similarity measurement becomes no longer applicable. A new similarity measurement was proposed by designing a high dimension FCM clustering algorithm. Firstly, an initialization of ordinal-numerical mappings is given; secondly, new ordinal-numerical mappings are learned from the iterative high dimension FCM clustering algorithm and the clustering effect becomes optimized at the same time; finally, a new similarity measurement of software cost data is proposed with the fuzzy partition matrix. The experimental results show that the similarity measurement improves the precision of software cost estimation.

Key words: software engineering, similarity measurement, high dimensionality, ordinal attribute

中图分类号: