[1]韩 磊,魏辉辉,徐 建.基于特征影响网络的类变更倾向度量及实证分析[J].南京理工大学学报(自然科学版),2013,37(06):839-844.
 Han Lei,Wei Huihui,Xu Jian.Empirical analysis and proneness metrics of class change based on feature influence network[J].Journal of Nanjing University of Science and Technology,2013,37(06):839-844.
点击复制

基于特征影响网络的类变更倾向度量及实证分析
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
37卷
期数:
2013年06期
页码:
839-844
栏目:
出版日期:
2013-12-31

文章信息/Info

Title:
Empirical analysis and proneness metrics of class change based on feature influence network
作者:
韩 磊1魏辉辉2徐 建2
1.华为技术有限公司 南京研究所,江苏 南京 210008;
2.南京理工大学 计算机科学与工程学院,江苏 南京 210094
Author(s):
Han Lei1Wei Huihui2Xu Jian2
1.Research and Development Center,Huawei Technologies Co.,Ltd.,Nanjing 210008,China;
2.School of Computer Science and Engineering,NUST,Nanjing 210094,China
关键词:
软件度量 软件变更 特征影响网络 面向对象软件 类变更
Keywords:
software metrics software change feature influence network object-oriented software class change
分类号:
TP393
摘要:
为了提高软件产品质量,降低软件变更维护成本,从软件网络视角设计有效的类变更趋势度量用于识别具有变更倾向的类。选取典型且类型各异的开源面向对象软件系统为研究对象,构造一种特征影响网络。结合特征影响网络的特点提出一种考虑软件变更直接影响关系的类变更趋势度量。实证分析结果表明该度量具备较好的类变更趋势预测能力,其最小匹配率高于60%,优于C&K度量中的加权方法数(WMC)、对象类耦合度(CBO)等指标。
Abstract:
To improve software product quality,reduce software change maintenance costs,valid class change proneness metrics is designed for recognizing classes with change proneness from the software network view.Taking typical and various types of open source object-oriented software as research objects,a feature influence network(FIN)is constructed.Class change proneness metric considering the direct influence relationship of software change is proposed combining the characteristics of the FIN.Empirical analysis results show that the metrics has good class proneness prediction capacity with the minimum matching rate of 60% and is better than weighted method count(WMC)and coupling between objects(CBO)of C&K metrics in terms of the matching rate.

参考文献/References:

[1] Singh Y,Kaur A,Malhotra R.Empirical validation of object-oriented metrics for predicting fault proneness models[J].Software Quality Journal,2010,18(1):3-35.
[2]Jie X,Ho D,Capretz L F.An empirical validation of object-oriented design metrics for fault prediction[J].Journal of Computer Sciences,2008,4(7):571-577.
[3]Gyimothy T,Ferenc R,Siket I.Empirical validation of object-oriented metrics on open source software for fault prediction[J].IEEE Transactions on Software Engineering,2005,31(10):897-910.
[4]易彤.面向对象设计中软件度量学:回顾与热点[J].计算机应用研究,2011,28(2):427-434.
Yi Tong.Software measurement study in object-oriented design:State of art[J].Application Research of Computers,2011,28(2):427-434.
[5]徐久强,刘红,赵海,等.软件网络中结构洞与紧密度的研究[J].东北大学学报(自然科学版),2010,31(11):1562-1565.
Xu Jiuqiang,Liu Hong,Zhao Hai,et al.Research on structural holes and closeness of software network[J].Journal of Northeastern University(Natural Science),2010,31(11):1562-1565.
[6]李兵,王浩,李增扬,等.基于复杂网络的软件复杂性度量研究[J].电子学报,2006(S1):2371-2375.
Li Bing,Wang Hao,Li Zhengyang,et al.Software complexity metrics based on complex networks[J].Acta Electronica Sinica,2006(S1):2371-2375.
[7]王树森,顾庆,陈焘,等.基于复杂网络的大型软件系统度量[J].计算机科学,2009,36(2):287-302.
Wang Shushen,Gu Qing,Chen Tao,et al.Metrics for large-scale software systems based on complex networks[J].Computer Science,2009,36(2):287-302.
[8]马于涛,何克清,李兵,等.网络化软件的复杂网络特性实证[J].软件学报,2011,22(3):381-407.
Ma Yutao,He Keqin,Li Bing,et al.Empirical study on the characteristics of complex networks in networked software[J].Journal of Software,2011,22(3):381-407.

备注/Memo

备注/Memo:
收稿日期:2013-06-18 修回日期:2013-08-20
基金项目:国家自然科学基金(61300053)
作者简介:韩磊(1979-),男,工程师,主要研究方向:新型程序设计方法,E-mail:h50466@huawei.com; 通讯作者:徐建(1979-),男,副教授,主要研究方向:可信软件,E-mail:dolphin.xu@njust.edu.cn。
引文格式:韩磊,魏辉辉,徐建.基于特征影响网络的类变更倾向度量及实证分析[J].南京理工大学学报,2013,37(6):839-844.
投稿网址:http://njlgdxxb.paperonce.org
更新日期/Last Update: 2013-12-31