Antirandom testing: getting the most out of black-box testing

Abstract

Random testing is a well known concept that requires that each test is selected randomly regardless of the test previously applied. This paper introduces the concept of antirandom testing. In this testing strategy each test applied is chosen such that its total distance from all previous tests is maximum. Two distance measures are de ned. Procedures to construct antirandom sequences are developed. A checkpoint encoding scheme is introduced that allows automatic generation of e cient test cases. Further developments and studies needed are identi ed.

Iterative Antirandom Testing

“…This information consists of the previous tests applied [14]. Therefore the new approach called Antirandom Testing (AT) was proposed.…”

“…Therefore the new approach called Antirandom Testing (AT) was proposed. In this case each test pattern is chosen such that its total distance from all previous patterns is maximum [14,25,26]. Distance is a measure of how different two test patterns T i and T j are [14].…”

“…In this case each test pattern is chosen such that its total distance from all previous patterns is maximum [14,25,26]. Distance is a measure of how different two test patterns T i and T j are [14]. So, to achieve higher fault coverage, one should choose test patterns that are as different as possible from the patterns previously generated.…”

“…For antirandom testing the Hamming Distance (HD) and Cartesian Distance (CD) have been used to measure the differences between two patterns [14,25].…”

Automatic test generation using checkpoint encoding and antirandom testing

“…There are many testing techniques discussed in the literature that can be termed black-box testing 6,8,9,13,20 . Random testing 8 chooses tests randomly based on some input distribution, without attempting to exploit information gained by tests applied earlier.…”