Learning to Represent and Understand Locative Prepositional Phrases
Cynthia L. Cosic, Paul Munro · D-Scholarship@Pitt (University of Pittsburgh) · 1988
The spatial juxtaposition observed between any two objects can generally be expressed by a locative prepositional phrase of the form noun-preposition-noun. A net work is described which learns to associate locative prepositions with the spatial relationships they express using back-propagation adaptation. Since the mapping is context sensitive, input patterns to the network include a representation of the objects: for example, the preposition on expresses different spatial relationships in the phrase "house on lake" and "plate on table". The network is designed such that, during the learning process, it must develop a small set of features for representing the nouns appropriately. The problem is framed as a pattern completion task in which the pattern to be completed consists of four components: two nouns, a preposition, and the spatial relationship; the missing component is either the preposition or the spatial relationship. After learning, the network was analyzed in terms of its ability to process novel pattern combinations, clustering of the noun representations, the core meanings of the prepositions, and the incremental influence of 0, 1, and 2 nouns on the interpreted meaning of a preposition.