Classification in Noisy Environments Using a Distance Measure Between Structural Symbolic Descriptions

A definition of distance measure between structural descriptions, which is based on a probabilistic interpretation of the matching predicate, is proposed. It aims at coping with the problem of classification when noise causes both local and structural deformations. The distance measure is defined according to a top-down evaluation scheme: distance between disjunctions of conjuncts, conjunctions, and literals. At the lowest level, the similarity between a feature value in the pattern model (G) and the corresponding value in the observation (Ez) is defined as the probability of observing a greater distortion. The classification problem is approached by means of a multilayered framework in which the cases of single perfect match, no perfect match, and multiple perfect match are treated differently. Another possible application of the distance measure is in the field of concept acquisition. A plausible solution for the problem of completing the attribute and structure spaces, based on the probabilistic approach, is also given. Finally, both a comparison with other related works and an application in the domain of layout-based document recognition are illustrated.