文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > Validating Object-Oriented Design Metrics on a Commercial Java Application
侵权投诉

Validating Object-Oriented Design Metrics on a Commercial Java Application

de l'information

Validating Object-Oriented

Design Metrics on a Commercial

Java Application

Daniela Glasberg1

Khaled El Emam2

Walcelio Melo3

Nazim Madhavji4

Abstract

Many of the object-oriented metrics that have been developed by the research community arebelieved to measure some aspect of complexity. As such, they can serve as leading indicators ofproblematic classes, for example, those classes that are most fault-prone. If faulty classes can bedetected early in the development project’s life cycle, mitigating actions can be taken, such asfocused inspections. Prediction models using design metrics can be used to identify faultyclasses early on. In this paper, we present a cognitive theory of object-oriented metrics and anempirical study which has as objectives to formally test this theory while validating the metricsand to build a post-release fault–proneness prediction model. The cognitive mechanisms whichwe apply in this study to object-oriented metrics are based on contemporary models of humanmemory. They are: familiarity, interference, and fan effects. Our empirical study was performedwith data from a commercial Java application. We found that Depth of Inheritance Tree (DIT) is agood measure of familiarity and, as predicted, has a quadratic relationship with fault–proneness.Our hypotheses were confirmed for Import Coupling to other classes, Export Coupling andNumber of Children metrics. The Ancestor based Import Coupling metrics were not associatedwith fault-proneness after controlling for the confounding effect of DIT. The prediction modelconstructed had a good accuracy. Finally, we formulated a cost savings model and applied it toour predictive model. This demonstrated a 42% reduction in post-release costs if the predictionmodel is used to identify the classes that should be inspected.

1 Introduction

A considerable number of object-oriented metrics have been developed by the researchcommunity. The basic premise behind these metrics is that they capture some elements ofobject-oriented software complexity. Examples of metrics include those defined in (Abreu andCarapuca, 1994; Benlarbi and Melo, 1999; Briand et al., 1997; Cartwright and Shepperd, 2000;Chidamber and Kemerer, 1994; Henderson-Sellers, 1996; Li and Henry, 1993; Lorenz and Kidd,1994; Tang et al., 1999). While many of these metrics are based on seemingly good ideas aboutwhat is important to measure in object-oriented software to capture its complexity, it is stillnecessary to empirically validate them. School of Computer Science McGill University, 3480 University Street, McConnell Engineering Building, Montreal,Quebec, Canada H3A 2A7. dglasb@cs.mcgill.ca

21 National Research Council of Canada, Institute for Information Technology, Building M-50, Montreal Road, Ottawa,Ontario, Canada K1A OR6. Khaled.El-Emam@nrc.ca

3

4 Oracle Brazil, SCN Qd. 2 Bl. A, Ed. Corporate, S. 604, 70712-900 Brasilia, DF, Brazil. wmelo@br.oracle.com School of Computer Science McGill University, 3480 University Street, McConnell Engineering Building, Montreal,Quebec, Canada H3A 2A7. madhavji@cs.mcgill.ca

上一页第3页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email