Modelling the Interpretation and Interpretation Ease of Noun-Noun Compounds Using a Relation Space Approach to Compound Meaning

In this paper, we present a computational model of conceptual combination that introduces a new representation for the meaning of compounds: the relations used to interpret compounds are represented as points or vectors in a high-dimensional relation space. Such a representational framework has many advantages over other approaches. Firstly, the high-dimensionality of the space provides a detailed description of the compound’s meaning; each of the space’s dimensions represents a semantically distinct way in which compound meanings can differ from each other. Secondly, the spatial representation allows for a distance metric to measure how similar of different pairs of compound meanings are to each other. We conducted a corpus study, generating vectors in this relation space representing the meanings of a large, representative set of familiar compounds. A computational model of compound interpretation that uses these vectors as a database from which to derive new relation vectors for new compounds is presented. Also presented is a model of interpretation ease: that is, the ease or rapidity with which people can comprehend compounds. Our model uses ideas from the CARIN theory of conceptual combination about the modifier noun’s role in the comprehension process; the model correlates as well as the traditional CARIN model with people’s reaction