A currently successful approach to computational semantics is to represent
words as embeddings in a machine-learned vector space. We present an ensemble
method that combines embeddings produced by GloVe (Pennington et al., 2014) and
word2vec (Mikolov et al., 2013) with structured knowledge from the semantic
networks ConceptNet (Speer and Havasi, 2012) and PPDB (Ganitkevitch et al.,
2013), merging their information into a common representation with a large,
multilingual vocabulary. The embeddings it produces achieve state-of-the-art
performance on many word-similarity evaluations. Its score of $= .596$ on
an evaluation of rare words (Luong et al., 2013) is 16% higher than the
previous best known system.
Beschreibung
[1604.01692] An Ensemble Method to Produce High-Quality Word Embeddings
%0 Generic
%1 speer2016ensemble
%A Speer, Robert
%A Chin, Joshua
%D 2016
%K embedding ensemble word
%T An Ensemble Method to Produce High-Quality Word Embeddings
%U http://arxiv.org/abs/1604.01692
%X A currently successful approach to computational semantics is to represent
words as embeddings in a machine-learned vector space. We present an ensemble
method that combines embeddings produced by GloVe (Pennington et al., 2014) and
word2vec (Mikolov et al., 2013) with structured knowledge from the semantic
networks ConceptNet (Speer and Havasi, 2012) and PPDB (Ganitkevitch et al.,
2013), merging their information into a common representation with a large,
multilingual vocabulary. The embeddings it produces achieve state-of-the-art
performance on many word-similarity evaluations. Its score of $= .596$ on
an evaluation of rare words (Luong et al., 2013) is 16% higher than the
previous best known system.
@misc{speer2016ensemble,
abstract = {A currently successful approach to computational semantics is to represent
words as embeddings in a machine-learned vector space. We present an ensemble
method that combines embeddings produced by GloVe (Pennington et al., 2014) and
word2vec (Mikolov et al., 2013) with structured knowledge from the semantic
networks ConceptNet (Speer and Havasi, 2012) and PPDB (Ganitkevitch et al.,
2013), merging their information into a common representation with a large,
multilingual vocabulary. The embeddings it produces achieve state-of-the-art
performance on many word-similarity evaluations. Its score of $\rho = .596$ on
an evaluation of rare words (Luong et al., 2013) is 16% higher than the
previous best known system.},
added-at = {2016-06-10T09:28:27.000+0200},
author = {Speer, Robert and Chin, Joshua},
biburl = {https://www.bibsonomy.org/bibtex/27b5ab7fcd68aa4e78e193c19c7bef601/thoni},
description = {[1604.01692] An Ensemble Method to Produce High-Quality Word Embeddings},
interhash = {b4b45c54c76486496e84d4eb27b6f05f},
intrahash = {7b5ab7fcd68aa4e78e193c19c7bef601},
keywords = {embedding ensemble word},
note = {cite arxiv:1604.01692},
timestamp = {2018-10-25T01:26:47.000+0200},
title = {An Ensemble Method to Produce High-Quality Word Embeddings},
url = {http://arxiv.org/abs/1604.01692},
year = 2016
}