首页 | 本学科首页   官方微博 | 高级检索  
     

基于代码修改的多目标有监督缺陷预测建模方法
引用本文:陈翔,王秋萍.基于代码修改的多目标有监督缺陷预测建模方法[J].计算机科学,2018,45(6):161-165.
作者姓名:陈翔  王秋萍
作者单位:南通大学计算机科学与技术学院 江苏 南通226019;南京大学软件新技术国家重点实验室 南京 210093;桂林电子科技大学广西可信软件重点实验室 广西 桂林541004,南通大学计算机科学与技术学院 江苏 南通226019
基金项目:本文受国家自然科学基金(61202006,61602267),南京大学计算机软件新技术国家重点实验室开放课题(KFKT2016B18),广西可信软件重点实验室研究课题(kx201610),江苏省高校自然科学研究项目(15KJB520030,16KJB520038),南通市科技平台项目(CP120130001)资助
摘    要:基于代码修改的缺陷预测,具有代码审查量少、缺陷定位和修复快的优点。文中首次将该问题建模为多目标优化问题,其中一个优化目标是最大化识别出的缺陷代码修改数,另一个优化目标是最小化需要审查的代码量。这两个优化目标之间存在一定的冲突,因此提出了MULTI方法,该方法可以生成一组具有非支配关系的预测模型。在实证研究中,考虑了6个大规模开源项目(累计227417个代码修改),以ACC和POPT作为评测预测性能的指标。实验结果表明,MULTI方法的预测性能均显著优于经典的有监督建模方法(EALR和Logistic)和无监督建模方法(LT和AGE)。

关 键 词:软件缺陷预测  多目标优化  代码修改  实证研究
收稿时间:2017/4/24 0:00:00
修稿时间:2017/7/19 0:00:00

Multi-objective Supervised Defect Prediction Modeling Method Based on Code Changes
CHEN Xiang and WANG Qiu-ping.Multi-objective Supervised Defect Prediction Modeling Method Based on Code Changes[J].Computer Science,2018,45(6):161-165.
Authors:CHEN Xiang and WANG Qiu-ping
Affiliation:School of Computer Science and Technology,Nantong University,Nantong,Jiangshu 226019,China;State Key Laboratory for Novel Software Technology at Nanjing University, Nanjing 210093,China;Guangxi Key Laboratory of Trusted Software,Guilin University of Electronic Technology,Guilin,Guangxi 541004,China and School of Computer Science and Technology,Nantong University,Nantong,Jiangshu 226019,China
Abstract:Defect prediction based on code changes has the advantage of smaller code inspection cost,easy fault localization and rapid fixing.This paper firstly formalized this problem as a multi-objective optimization problem.One objective is to maximize the number of identified buggy changes,and the other objective is to minimize the cost of code inspection.There exist an obvious conflict between two objectives,so this paper proposed a novel method MULTI.This me-thod can generate a set of non-dominated prediction models.In the empirical studies,this paper chose six large-scale open source projects (with 227417 code changes in total) and considerd ACC and POPT as evaluation indicators of perfor-mance.Final results show that the proposed method can perform significantly better than the state-of-the-art supervised methods (i.e.,EALR and Logistic) and unsupervised methods (i.e.,LT and AGE).
Keywords:Software defect prediction  Multi-objective optimization  Code changes  Empirical studies
点击此处可从《计算机科学》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号