Intelligent versus Random Software Testing

Juichi TAKAHASHI  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E87-D   No.4   pp.849-854
Publication Date: 2004/04/01
Online ISSN: 
DOI: 
Print ISSN: 0916-8532
Type of Manuscript: Special Section PAPER (Special Section on Knowledge-Based Software Engineering)
Category: Metrics, Test, and Maintenance
Keyword: 
software testing,  random testing,  intelligence testing,  Chinese postman problem,  fault detection,  

Full Text: PDF(346.5KB)>>
Buy this Article




Summary: 
Comparison of intelligent and random testing in data inputting is still under discussion. Little is also known about testing for the whole software and empirical testing methodology when random testing used. This study research not only for data inputting testing, but also operation of software (called transitions) in order to test the whole GUI software by intelligent and random testing. Methodology of this study is that we attempt to research efficiency of random and intelligent testing by Chinese postman problem. In general, random testing is considered straightforward but not efficient. Chinese postman problem testing is complicated but efficient. The comparison between random and intelligent testing would give further recommendation for software testing methodology.