Mixed Diagnostic Tests Buildingfrom an Optimal Unconditional Test Subsetin Intelligent Pattern Recognition

摘要

In intelligent pattern recognition, it is proposed that mixed diagnostic test (MDT) building beconducted from an optimal subset (OS) of unconditional irredundant diagnostic tests (UIDT). Decisionmaking on each MDT is performed in the course of MDT tree building. The criteria of considering thesequence for UIDT as well as features involved in each of them are stated, which results in enumerationreduction and, as a rule, allows MDT building from not all UIDT in an UIDT OS. A matrix model of dataand knowledge representation, a method and an algorithm of the MDT tree building, an illustrative example,and the ways of algorithm development are cited.