Inducing Fine-Grained Semantic Classes via Hierarchical and Collective Classification

摘要

Research in named entity recognition and mention detection has typically involved a fairly small number of semantic classes, which may not be adequate if semantic class information is intended to support natural language applications. Motivated by this observation, we examine the under-studied problem of semantic subtype induction, where the goal is to automatically determine which of a set of 92 fine-grained semantic classes a noun phrase belongs to. We seek to improve the standard supervised approach to this problem using two techniques: hierarchical classification and collective classification. Experimental results demonstrate the effectiveness of these techniques, whether or not they are applied in isolation or in combination with the standard approach.