SHAPE: a machine learning system from examples

This paper presents a new machine learning system called SHAPE. The input data are vectors of properties (represented as attribute-value pairs) which are used to describe individual cases, examples or observations in a given world. Each case belongs to exactly one of a set of classes, and the aim is to produce a collection of decision rules concluding the class according to the properties observed. SHAPE follows three steps. First, we build up an acyclic graph capturing dependencies among the properties involved. Since we endow this net with a semantic interpretation, we are allowed to read the net as a first draft of classification rules. Secondly, we will rewrite these rules to compact their syntactic description using automata minimization techniques. Finally, the rules are generalized in order to obtain the definite intensional description of the thus learned concepts. Let us remark that the last two stages could be applied to a set of rules attached to a collection of examples coming from any other learning system. To close the paper, we also present different experiments made with SHAPE to illustrate the performance of the system in a wide range of applications.