TY - GEN
T1 - A random and coverage-based approach for fault localization prioritization
AU - Zhang, Xiao Yi
AU - Towey, Dave
AU - Chen, Tsong Yueh
AU - Zheng, Zheng
AU - Cai, Kai Yuan
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/8/3
Y1 - 2016/8/3
N2 - Fault Localization Prioritization (FLP) aims to order the execution sequence of test cases so that faulty statements in a faulty program can be localized faster. FLP is an important part of the automation of testing and fault localization in software engineering. The key issue is to identify which test cases can provide most useful information to help locate the faulty statement. Assuming the well-known technique of Spectra-Based Fault Localization (SBFL) is applied, this paper evaluates the quality of a test case based on the characteristics of its statement coverage information. We propose the COverage-based Random (COR) approach to address the FLP problem. Two statement coverage characteristics, the diversity characteristic and the failure-like characteristic, are analyzed and identified as having significant impacts on the effectiveness of fault localization. When using the COR approach, each test case is examined and the degree of each characteristic is measured, with test cases showing high degrees of the characteristics being assigned higher priority for execution. Because of the power of random strategies to improve the robustness of the approach, some random factors in the selection of test cases are included. Empirical studies show that, compared with existing approaches, the COR approach results in a faster localization of faulty statements, reducing the number of necessary test case executions.
AB - Fault Localization Prioritization (FLP) aims to order the execution sequence of test cases so that faulty statements in a faulty program can be localized faster. FLP is an important part of the automation of testing and fault localization in software engineering. The key issue is to identify which test cases can provide most useful information to help locate the faulty statement. Assuming the well-known technique of Spectra-Based Fault Localization (SBFL) is applied, this paper evaluates the quality of a test case based on the characteristics of its statement coverage information. We propose the COverage-based Random (COR) approach to address the FLP problem. Two statement coverage characteristics, the diversity characteristic and the failure-like characteristic, are analyzed and identified as having significant impacts on the effectiveness of fault localization. When using the COR approach, each test case is examined and the degree of each characteristic is measured, with test cases showing high degrees of the characteristics being assigned higher priority for execution. Because of the power of random strategies to improve the robustness of the approach, some random factors in the selection of test cases are included. Empirical studies show that, compared with existing approaches, the COR approach results in a faster localization of faulty statements, reducing the number of necessary test case executions.
KW - COverage-based approaches
KW - software fault localization
KW - test case prioritization
UR - http://www.scopus.com/inward/record.url?scp=84983735886&partnerID=8YFLogxK
U2 - 10.1109/CCDC.2016.7531562
DO - 10.1109/CCDC.2016.7531562
M3 - Conference contribution
AN - SCOPUS:84983735886
T3 - Proceedings of the 28th Chinese Control and Decision Conference, CCDC 2016
SP - 3354
EP - 3361
BT - Proceedings of the 28th Chinese Control and Decision Conference, CCDC 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 28th Chinese Control and Decision Conference, CCDC 2016
Y2 - 28 May 2016 through 30 May 2016
ER -