PURPOSE: To learn the search control knowledge in an inference process from experience in solving a problem for proving a conclusion problem element from ≥1 premise conditions. ;CONSTITUTION: This system consists of a general knowledge base 1, a search control knowledge base 2, a recognition relation knowledge base 3, an input device 4, an output device 5, an inference processor 6, and a temporary work memory area 7. The knowledge stored in the search control knowledge base 2 suggests in what state and in what order the knowledge stored in the general knowledge base 1 is used and the recognition relation knowledge base 3 stores knowledge wherein conditions as to whether an area object can be recognized are described. A problem inputted from the input device 4 is solved and the search control knowledge is learnt from its problem history; and the result is outputted to the output device 5 and also stored in the search control knowledge base 2.;COPYRIGHT: (C)1995,JPO
展开▼
