In this paper we present an algorithm for learning a function able to assess objects. Our approach assumes that the experts can provide a collection of pair-wise comparisons. However, they have difficulties in assigning a number to the qualities of the objects considered. This is a typical situation when dealing with food products, where it is very interesting to have repeatable, reliable mechanisms that are as objective as possible to evaluate quality in order to provide markets with products of a uniform quality. The algorithm is implemented using a growing variant of Kohonen's Self-Organizing Maps (growing neural gas). Its performance has been tested with a variety of artificial or public data sets to demonstrate the capabilities of our approach.