An Experimental Study on the Interpretability of Fuzzy Systems

摘要: Interpretability is one of the most significant proper- ties Fuzzy Systems which are widely acknowledged as gray boxes against other Soft Computing techniques such Neural Networks usually regarded black boxes. It essential for applications with high human interaction (decision support systems in medicine, eco- nomics, etc). The use accuracy indices to guide fuzzy modeling process broadly extended. In turn, although there have been a few attempts define indices, we still far away from having universal index. With aim evaluating used an experimental analysis (in form web poll) was carried out yielding some useful clues keep mind regard- ing assessment. Results extracted poll show inherent subjectivity measure because collected huge diversity answers. Nevertheless, comparing carefully all an- swers, it possible find interesting user profiles. Keywords— modeling,