ABSTRACT
Another weakness of the current approach to verification and validation testing is the number and selection of test cases used. Typically, the number has been very small; for example, only 10 test cases were utilized for the validation of MYCIN, and 50 for Rl. Additionally, the selection of the test cases has been unsystematic. Test cases are chosen without regard to their representation of the problem or knowledge domain. The selection process also does not consider the coverage of test cases. Coverage is the measure ment of how effectively the test cases verify or validate the expert system.
