Abstract

BAKAROV, Amir. Distributional Word Vectors as Semantic Maps Framework. Comp. y Sist. [online]. 2022, vol.26, n.3, pp.1343-1364.  Epub Dec 02, 2022. ISSN 2007-9737.  https://doi.org/10.13053/cys-26-3-4356.

Distributional Semantics Models are one of the most ubiquitous tools in Natural Language Processing. However, it is still unclear how to optimize such models for specific tasks and how to evaluate them in a general setting (having ability to be successfully applied to any language task in mind). We argue that benefits of intrinsic distributional semantic models evaluation could be questioned since the notion of their “general quality” possibly does not exist; distributional semantic models, however, can be considered as a part of Semantic Maps framework which formalizes the notion of linguistic representativeness on the lexical level.

Keywords : Word embeddings; distributional word vectors; semantic maps.