15.ニューラルネットワークチューニングによるファジィ推論尺(ファジィニューラル(2),一般講演,医療とヒューマンフレンドリーコンピュータ科学の接点を求めて-21世紀マルチメディア医療をむかえて-)

抄録

We proposed a fuzzy inference ruler using d'Ocaone's nomography in the meeting of BMFSA 1994. We were able to get results of fuzzy inference by the ruler without computer. However , it was difficult to make a membership function. So we propose a fuzzy inference ruler using neuro-learning method. We will be able to get a membership function automatically by using the method. It is expected that the fuzzy inference ruler using neuro-learning will be accepted and utilized. In this paper , we will first refer to the nomogram again, then discuss how to use neuro-learning method in the fuzzy inference ruler.