Paper: Supporting inferences in semantic space: representing words as regions

ACL ID W09-3711
Title Supporting inferences in semantic space: representing words as regions
Venue International Conference on Computational Semantics
Session Main Conference
Year 2009
Authors
  • Katrin Erk (University of Texas at Austin, Austin TX)

Semantic space models represent the meaning of a word as a vector in high-dimensional space. They offer a framework in which the mean- ing representation of a word can be computed from its context, but the question remains how they support inferences. While there has been some work on paraphrase-based inferences in semantic space, it is not clear how semantic space models would support inferences involving hyponymy, like horse ran → animal moved. In this paper, we first dis- cuss what a point in semantic space stands for, contrasting semantic space with G¨ardenforsian conceptual space. Building on this, we pro- pose an extension of the semantic space representation from a point to a region. We present a model for learning a region representation for word meaning in semantic space, based...