Описан способ усовершенствования модели многомерного нечеткого распознавателя, позволяющий исключить основной недостаток, присущий модели, реализуемой на основе методов типа Мамдани.
A resolver model that produces resolver and creates opportunities advantages over methods such Mamdani, which has the disadvantage that any object is classified as a clear, even if it belongs to a corresponding cluster is questionable is described. Improvement of the model recognizer with a focus on the issue of fuzzy measure values accessories when needed, it is shown in this paper is grounded.
The authors declare that there are no conflicts of interest present.