Method of the Debugging of the Knowledge Bases of Intellectual Decision Making Systems
Olga N. Dolinina · Advances in intelligent systems and computing · 2016
The paper describes the method of the debugging of the intellectual decision making systems. It combines the detecting of the structural errors and the testing of the knowledge base. Static analysis allows to detect so called structural errors such as incomplete knowledge, inconsistency, extra rules. Static debugging allows to build the static correct knowledge base. But even the static correct knowledge base can have errors connected with the inconsistency of the subject area which can be detected with the dynamic debugging (testing). The paper shows that the most difficult for the detection is the “forgetting about the exception” type of the errors. There is described the method of the generation of the full test set which allows to detect such types of the errors in the knowledge base. The method is based on the building of the tests for the logic schemes. The method was successfully approved for the testing of the rule-based expert systems and for the artificial network based on the 3-level perceptron.