On the selection of strength for fixed-strength interaction coverage based prioritization

Abstract test cases are derived by modeling the system under test, and have been widely applied in practice, such as for software product line testing and combinatorial testing. Abstract test case prioritization (ATCP) is used to prioritize abstract test cases and aims at achieving higher rates of f...

Full description

Bibliographic Details
Main Authors: Huang, Rubing, Zong, Weiwen, Chen, Tsong Yueh, Towey, Dave, Chen, Jinfu, Zhou, Yunan, Sun, Weifeng
Format: Conference or Workshop Item
Language:English
Published: 2018
Subjects:
Online Access:https://eprints.nottingham.ac.uk/55541/
_version_ 1848799179600560128
author Huang, Rubing
Zong, Weiwen
Chen, Tsong Yueh
Towey, Dave
Chen, Jinfu
Zhou, Yunan
Sun, Weifeng
author_facet Huang, Rubing
Zong, Weiwen
Chen, Tsong Yueh
Towey, Dave
Chen, Jinfu
Zhou, Yunan
Sun, Weifeng
author_sort Huang, Rubing
building Nottingham Research Data Repository
collection Online Access
description Abstract test cases are derived by modeling the system under test, and have been widely applied in practice, such as for software product line testing and combinatorial testing. Abstract test case prioritization (ATCP) is used to prioritize abstract test cases and aims at achieving higher rates of fault detection. Many ATCP algorithms have been proposed, using different prioritization criteria and information. One ATCP approach makes use of fixed-strength level-combinations information covered by abstract test cases, and is called fixed-strength interaction coverage based prioritization (FICBP). Before using FICBP, the prioritization strength λ needs to be decided. Previous studies have generally focused on λ values ranging between 1 and 6. However, no study has investigated the appropriateness of such a range, nor how to assign the prioritization strength for FICBP. To answer these questions, this paper reports on an empirical study involving four real-life programs (each of which with six versions). The experimental results indicate that λ should be set approximately equal to a value corresponding to half of the number of parameters, when testing resources are sufficient. Our results also show that when testing resources are limited or insufficient, either small or large λ values are suggested for FICBP.
first_indexed 2025-11-14T20:31:33Z
format Conference or Workshop Item
id nottingham-55541
institution University of Nottingham Malaysia Campus
institution_category Local University
language English
last_indexed 2025-11-14T20:31:33Z
publishDate 2018
recordtype eprints
repository_type Digital Repository
spelling nottingham-555412018-11-15T13:56:01Z https://eprints.nottingham.ac.uk/55541/ On the selection of strength for fixed-strength interaction coverage based prioritization Huang, Rubing Zong, Weiwen Chen, Tsong Yueh Towey, Dave Chen, Jinfu Zhou, Yunan Sun, Weifeng Abstract test cases are derived by modeling the system under test, and have been widely applied in practice, such as for software product line testing and combinatorial testing. Abstract test case prioritization (ATCP) is used to prioritize abstract test cases and aims at achieving higher rates of fault detection. Many ATCP algorithms have been proposed, using different prioritization criteria and information. One ATCP approach makes use of fixed-strength level-combinations information covered by abstract test cases, and is called fixed-strength interaction coverage based prioritization (FICBP). Before using FICBP, the prioritization strength λ needs to be decided. Previous studies have generally focused on λ values ranging between 1 and 6. However, no study has investigated the appropriateness of such a range, nor how to assign the prioritization strength for FICBP. To answer these questions, this paper reports on an empirical study involving four real-life programs (each of which with six versions). The experimental results indicate that λ should be set approximately equal to a value corresponding to half of the number of parameters, when testing resources are sufficient. Our results also show that when testing resources are limited or insufficient, either small or large λ values are suggested for FICBP. 2018-06-18 Conference or Workshop Item PeerReviewed application/pdf en https://eprints.nottingham.ac.uk/55541/1/HuangZongChenToweyZhouChen.COMPSAC2016.AuthorCopy.pdf Huang, Rubing, Zong, Weiwen, Chen, Tsong Yueh, Towey, Dave, Chen, Jinfu, Zhou, Yunan and Sun, Weifeng (2018) On the selection of strength for fixed-strength interaction coverage based prioritization. In: IEEE Computer Society's International Computer Software and Applications Conference, 23-27 Jul 2018, Tokyo, Japan. Software testing; abstract test case prioritization; FICBP; prioritization strength; empirical study https://ieeexplore.ieee.org/document/8377673 10.1109/COMPSAC.2018.00049 10.1109/COMPSAC.2018.00049 10.1109/COMPSAC.2018.00049
spellingShingle Software testing; abstract test case prioritization; FICBP; prioritization strength; empirical study
Huang, Rubing
Zong, Weiwen
Chen, Tsong Yueh
Towey, Dave
Chen, Jinfu
Zhou, Yunan
Sun, Weifeng
On the selection of strength for fixed-strength interaction coverage based prioritization
title On the selection of strength for fixed-strength interaction coverage based prioritization
title_full On the selection of strength for fixed-strength interaction coverage based prioritization
title_fullStr On the selection of strength for fixed-strength interaction coverage based prioritization
title_full_unstemmed On the selection of strength for fixed-strength interaction coverage based prioritization
title_short On the selection of strength for fixed-strength interaction coverage based prioritization
title_sort on the selection of strength for fixed-strength interaction coverage based prioritization
topic Software testing; abstract test case prioritization; FICBP; prioritization strength; empirical study
url https://eprints.nottingham.ac.uk/55541/
https://eprints.nottingham.ac.uk/55541/
https://eprints.nottingham.ac.uk/55541/