可缓解类重叠问题的跨版本软件缺陷预测方法  

Cross-Version Software Defect Prediction Methodfor Relieving Class Overlap Problem

作　　者：曲豫宾 陈翔[3] 李龙[1] QU Yubin;CHEN Xiang;LI Long(Guangxi Key Laboratory of Trusted Software,Guilin University of Electronic Technology,Guilin 541004,Guangxi Zhuang Autonomous Region,China;School of Information Engineering,Jiangsu College of Engineering and Technology,Nantong 226001,Jiangsu Province,China;School of Information Science and Technology,Nantong University,Nantong 226019,Jiangsu Province,China)

机构地区：[1]桂林电子科技大学广西可信软件重点实验室,广西桂林541004 [2]江苏工程职业技术学院信息工程学院,江苏南通226001 [3]南通大学信息科学技术学院,江苏南通226019

出　　处：《吉林大学学报（理学版）》2021年第2期372-378,共7页Journal of Jilin University:Science Edition

基　　金：国家自然科学基金青年科学基金(批准号:61202006);南通市科技计划指令性项目(批准号:JC2019106);广西可信软件重点实验室研究项目(批准号:kx202013);江苏工程职业技术学院科研计划项目(批准号:GYKY/2020/4);江苏高校“青蓝工程”项目.

摘　　要：针对软件缺陷预测过程中未充分使用源代码语义特征以及训练数据集中的类重叠问题,提出一种面向类重叠的跨版本软件缺陷深度特征学习方法.该方法采用混合式最近邻清理策略缓解深度学习语义特征中存在的类重叠问题.在PROMISE公开数据集上进行测试的结果表明,该策略能提升基于深度语义学习的软件缺陷预测性能,分类性能最多在中值上提升14.8%.实验结果表明,在跨版本深度缺陷预测问题中可采用混合式最近邻清理策略缓解类重叠问题.Aiming at the problem that semantic features of source code were not fully used in the process of software defect prediction and class overlap in training data set,we proposed a cross-version software defect deep feature learning method for class overlap.This method used a hybrid nearest neighbor cleaning strate gy to alleviate class overlap problem in deep learning semantic features.The test results on open data set PROMISE show that this strategy can improve the performance of software defect prediction based on de ep semantic learning,and the classification performance can be improved by 14.8%at most in the median value.The experimental results show that a hybrid nearest neighbor cleaning strategy can be used to alleviate the class overlap problem in the cross-version deep defect prediction problem.

关 键 词：软件缺陷预测 深度学习 类重叠 语义特征 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]