In real applications, data provided to a learning system usually contain li
nguistic information which greatly influences concept descriptions derived
by conventional inductive learning methods. The design of learning methods
to learn concept descriptions in working with vague data is thus very impor
tant. In this paper, we apply fuzzy set concepts to machine learning to sol
ve this problem. A fuzzy learning algorithm based on the maximum informatio
n gain is proposed to manage linguistic information. The proposed learning
algorithm generates fuzzy rules from "soft" instances, which differ from co
nventional instances in that they have class membership values. Experiments
on the Sports and the Iris Flower classification problems are presented to
compare the accuracy of the proposed algorithm with those of some other le
arning algorithms. Experimental results show that the rules derived from ou
r approach are simpler and yield higher accuracy than those from some other
learning algorithms. (C) 1999 Elsevier Science B.V. All rights reserved.