FUZZY CONTROL OF VEHICLE INTERPRETATION BASED ON RELATION BETWEEN GUILFORD'S CUBE AND KOSKO'S CUBE
 
Gheorghe, P.T., West University of Timisoara, Romania
 
Fuzzy decisions networks can be constructed from training examples by machine learning techniques, and the connectionist structure can be trained to develop fuzzy logic rule and find optimal input/output membership function. This model also provides human-understandable meaning to the normal feedforward multilayer neural network in witch the internal units are always opaque to the users. For interpretation the results with linguistic variables was necessaries relating two concepts: the Guilford's cube for representation the items of intelligence and the Kosko's cube for representation the fuzzy numbers. Two examples are presented to illustrate the performance and applicability of the proposed connectionist model in psychology. In first example, the proposed connectionist model for a fuzzy logic controller was used to simulate the control of the subjects working in the domaines with high-performance (Traffic and Transport). In the second example, the proposed connectionist model for a fuzzy logic decision- making system was used to decide a proper scheduling based on some chosen items in intelligence tests.