JCSE, vol. 2, no. 3, pp.274-300, 2008

DOI:

An Empirical Evaluation of Test Data Generation Techniques

Seung-Hee Han Yong-Rae Kwon
Network Technology Laboratory, KT, Korea|Division of Computer Science, Dept. EECS, KAIST, Korea

Abstract: Software testing cost can be reduced if the process of testing is automated. However, the testdata generation task is still performed mostly by hand although numerous theoretical workshave been proposed to automate the process of generating test data and even commercial testdata generators appeared on the market. Despite prolific research reports, few attempts havebeen made to evaluate and characterize those techniques. Therefore, a lot of works have beenproposed to automate the process of generating test data. However, there is no overallevaluation and comparison of these techniques. Evaluation and comparison of existingtechniques are useful for choosing appropriate approaches for particular applications, and alsoprovide insights into the strengths and weaknesses of current methods. This paper conductsexperiments on four representative test data generation techniques and discusses theexperimental results. The results of the experiments show that the genetic algorithm (GA)-based test data generation performs the best. However, there are still some weaknesses in theGA-based method. Therefore, we modify the standard GA-based method to cope with theseweaknesses. The experiments are carried out to compare the standard GA-based method andtwo modified versions of the GA-based method.

Keyword: No keyword

Full Paper:   141 Downloads, 3666 View