Embedding a Semantic Network in a Word Space

Human Language Technologies: The 2015 Annual Conference of the North American Chapter of the ACL, pages 1428–1433, Denver, Colorado, May 31 – June 5, 2015. c⃝2015 Association for Computational Linguistics Embedding a Semantic Network in a Word Space Richard Johansson and Luis Nieto Pi˜na Spr˚akbanken, Department of Swedish, University of Gothenburg Box 200, SE-40530 Gothenburg, Sweden {richard.johansson, luis.nieto.pina}@svenska.gu.se Abstract We present a framework for using continuous- space vector representations of word meaning to derive new vectors representing the mean- ing of senses listed in a semantic network. It is a post-processing approach that can be ap- plied to several types of word vector represen- tations. It uses two ideas: first, that vectors for polysemous words can be decomposed into a convex combination of sense vectors; sec- ondly, that the vector for a sense is kept sim- ilar to those of its neighbors in the network. This leads to a constrained optimization prob- lem, and we present an approximation for the case when the distance function is the squared Euclidean. We applied this algorithm on a Swedish se- mantic network, and we evaluate the quality of the resulting sense representations extrinsi- cally by showing that they give large improve- ments when used in a classifier that creates lexical units for FrameNet frames. 1 Introduction Representing word meaning computationally is cen- tral in natural language processing. Manual, knowledge-based approaches to meaning represen- tation maps word strings to symbolic concepts, which can be described using any knowledge rep- resentation framework; using the relations between concepts defined in the knowledge base, we can in- fer implicit facts from the information stated in a text: a mouse is a rodent, so it has prominent teeth. Conversely, data-driven meaning representation approaches rely on cooccurrence patterns to derive a vector representation (Turney and Pantel, 2010). There are two classes of methods that compute word vectors: context-counting and context-predicting; while the latter has seen much interest lately, their respective strengths and weaknesses are still being debated (Baroni et al., 2014; Levy and Goldberg, 2014). The most important relation defined in a vec- tor space between the meaning of two words is sim- ilarity: a mouse is something quite similar to a rat. Similarity of meaning is operationalized in terms of geometry, by defining a distance metric. Symbolic representations seem to have an advan- tage in describing word sense ambiguity: when a surface form corresponds to more than one concept. For instance, the word mouse can refer to a rodent or an electronic device. Vector-space representations typically represent surface forms only, which makes it hard to search e.g. for a group of words similar to the rodent sense of mouse or to reliably use the vectors in classifiers that rely on the semantics of the word. There have been several attempts to create vectors representing senses, most of them based on some variant of the idea first proposed by Sch¨utze (1998): that senses can be seen as clusters of similar contexts. Recent examples in this tradition include the work …

Content truncated to 500 words