Open AccessTest Data Generation for Multiple Paths Basedon Local Evolution
Author(s) -
Yao Xiangjuan,
Gong Dunwei,
Wang Wenliang
Publication year - 2015
Publication title -
chinese journal of electronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.267
H-Index - 25
eISSN - 2075-5597
pISSN - 1022-4653
DOI - 10.1049/cje.2015.01.008
Subject(s) - granularity , path (computing) , computer science , computation , test data , algorithm , test (biology) , data mining , programming language , operating system , paleontology , biology
Generating test data by genetic algorithmsis a promising research direction in software testing, amongwhich path coverage is an important test method. Theefficiency of test data generation for multi‐path coverageneeds to be further improved. We propose a test data generationmethod for multi‐path coverage based on a geneticalgorithm with local evolution. The mathematical model isestablished for all target paths, while in the algorithm theindividuals are evolved locally according to different objectivefunctions. We can improve the utilization efficiency oftest data. The computation cost can be reduced by usingfitness functions of different granularity in different phasesof the algorithm.