Fuzzy quantifiers and quantifying operators in a connectionistexpert system development tool

摘要

The authors give information pertaining to the implementation ofnfuzzy quantifiers and quantifying operators within a connectionistnnetwork model. The operators described can be extended to arbitraryninput size, by retaining similar overall behavior. Examples are given tonshow the responses one would obtain when modifying the belief in theninputs. These outputs correspond to responses one would intuitivelynexpect. The importance of having been able to implement these constructsnis given in the possibility of formulating more natural-language-likenconstructs. In conjunction with learning, complex symbolic systems maynbe modeled in domains which contain significant imprecision in anconnectionist network using the technique considered. The methods couldnalso be adapted to other connectionist architectures