Fuzzy quantifiers and OWA operators in inductive learning from erroneous examples

We propose the use of Yager's (1988) ordered weighted averaging (OWA) operators as a means for fuzzy linguistic quantifier based aggregation in inductive learning (from examples) under errors, misclassifications, etc. To somehow neglect those errors, we formulate the problem so as to find a concept description covering, say, almost all of the positive examples and almost none of the negative examples. Thus, by neglecting some examples, those errors are somehow "masked".