Over the past years, distributed representations have proven effective and
flexible keepers of prior knowledge to be integrated into downstream
applications. This survey is focused on semantic representation of meaning. We
start from the theoretical background behind word vector space models and
highlight one of their main limitations: the meaning conflation deficiency,
which arises from representing a word with all its possible meanings as a
single vector. Then, we explain how this deficiency can be addressed through a
transition from word level to the more fine-grained level of word senses (in
its broader acceptation) as a method for modelling unambiguous lexical meaning.
We present a comprehensive overview of the wide range of techniques in the two
main branches of sense representation, i.e., unsupervised and knowledge-based.
Finally, this survey covers the main evaluation procedures and provides an
analysis of five important aspects: interpretability, sense granularity,
adaptability to different domains, compositionality and integration into
downstream applications.
Description
[1805.04032] From Word to Sense Embeddings: A Survey on Vector Representations of Meaning
%0 Generic
%1 camachocollados2018sense
%A Camacho-Collados, Jose
%A Pilehvar, Mohammad Taher
%D 2018
%K embedding senseembedding wsd
%T From Word to Sense Embeddings: A Survey on Vector Representations of
Meaning
%U http://arxiv.org/abs/1805.04032
%X Over the past years, distributed representations have proven effective and
flexible keepers of prior knowledge to be integrated into downstream
applications. This survey is focused on semantic representation of meaning. We
start from the theoretical background behind word vector space models and
highlight one of their main limitations: the meaning conflation deficiency,
which arises from representing a word with all its possible meanings as a
single vector. Then, we explain how this deficiency can be addressed through a
transition from word level to the more fine-grained level of word senses (in
its broader acceptation) as a method for modelling unambiguous lexical meaning.
We present a comprehensive overview of the wide range of techniques in the two
main branches of sense representation, i.e., unsupervised and knowledge-based.
Finally, this survey covers the main evaluation procedures and provides an
analysis of five important aspects: interpretability, sense granularity,
adaptability to different domains, compositionality and integration into
downstream applications.
@misc{camachocollados2018sense,
abstract = {Over the past years, distributed representations have proven effective and
flexible keepers of prior knowledge to be integrated into downstream
applications. This survey is focused on semantic representation of meaning. We
start from the theoretical background behind word vector space models and
highlight one of their main limitations: the meaning conflation deficiency,
which arises from representing a word with all its possible meanings as a
single vector. Then, we explain how this deficiency can be addressed through a
transition from word level to the more fine-grained level of word senses (in
its broader acceptation) as a method for modelling unambiguous lexical meaning.
We present a comprehensive overview of the wide range of techniques in the two
main branches of sense representation, i.e., unsupervised and knowledge-based.
Finally, this survey covers the main evaluation procedures and provides an
analysis of five important aspects: interpretability, sense granularity,
adaptability to different domains, compositionality and integration into
downstream applications.},
added-at = {2018-09-17T22:39:41.000+0200},
author = {Camacho-Collados, Jose and Pilehvar, Mohammad Taher},
biburl = {https://www.bibsonomy.org/bibtex/2cd499bb19b6820a2bddbabad5925aa53/albinzehe},
description = {[1805.04032] From Word to Sense Embeddings: A Survey on Vector Representations of Meaning},
interhash = {0259f6f91a6754a4c3e071c9a40ee156},
intrahash = {cd499bb19b6820a2bddbabad5925aa53},
keywords = {embedding senseembedding wsd},
note = {cite arxiv:1805.04032Comment: 40 pages, 8 figures. Submitted to JAIR},
timestamp = {2018-09-17T22:39:41.000+0200},
title = {From Word to Sense Embeddings: A Survey on Vector Representations of
Meaning},
url = {http://arxiv.org/abs/1805.04032},
year = 2018
}