In this paper, a data mining process to induce a set of fuzzy rules from a database is presented. This process is based on the construction of fuzzy decision trees. We present a method to construct fuzzy decision trees and a method to use them to classify new examples. In presence of databases, prerequisites for training sets are introduced to generate a good subset of data that will enable us to construct a fuzzy decision tree. Moreover, we present different kinds of rules that can be induced by means of the construction of a decision tree, and we discuss some possible uses of such knowledge.