Markov chain Monte Carlo random testing

Authors:
Bo Zhou;Hiroyuki Okamura;Tadashi Dohi
Affiliations:
Department of Information Engineering, Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, Japan;Department of Information Engineering, Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, Japan;Department of Information Engineering, Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, Japan
Venue:
AST/UCMA/ISA/ACN'10 Proceedings of the 2010 international conference on Advances in computer science and information technology
Year:
2010

Citing 6
Cited 0

Automated Software Test Data Generation

IEEE Transactions on Software Engineering
The chaining approach for software test data generation

ACM Transactions on Software Engineering and Methodology (TOSEM)
Proportional sampling strategy: a compendium and some insights

Journal of Systems and Software
An Innovative Approach to Tackling the Boundary Effect in Adaptive Random Testing

HICSS '07 Proceedings of the 40th Annual Hawaii International Conference on System Sciences
Normalized restricted random testing

Ada-Europe'03 Proceedings of the 8th Ada-Europe international conference on Reliable software technologies
Adaptive random testing

ASIAN'04 Proceedings of the 9th Asian Computing Science conference on Advances in Computer Science: dedicated to Jean-Louis Lassez on the Occasion of His 5th Cycle Birthday

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a software random testing scheme based on Markov chain Monte Carlo (MCMC) method. The significant issue of software testing is how to use the prior knowledge of experienced testers and the information obtained from the preceding test outcomes in making test cases. The concept of Markov chain Monte Carlo random testing (MCMCRT) is based on the Bayes approach to parametric models for software testing, and can utilize the prior knowledge and the information on preceding test outcomes for their parameter estimation. In numerical experiments, we examine effectiveness of MCMCRT with ordinary random testing and adaptive random testing.