Knowledge Acquisition with Deep Fuzzy Inference Model and Its Application to a Medical Diagnosis

摘要: In this paper, we reduce the number of fuzzy rules in inference model and acquire knowledge as rules. The input items used for is reduced by randomly selecting each layer. Therefore, it turns out that whole can be more than an uses all original at one time. However, previous Zhang, although consequent part rule was learned, antecedent not learned. Since need to deal with situation where there no prior problem apply will necessary from data, required learn part. propose a learning method sets order obtain relationship between output data actual data. Then, example, proposed applied medical diagnosis diabetes, accuracy compared method.