Tailoring Word Embeddings for Bilexical Predictions: An Experimental Comparison
Computation and Language
2015-04-13 v2 Machine Learning
Abstract
We investigate the problem of inducing word embeddings that are tailored for a particular bilexical relation. Our learning algorithm takes an existing lexical vector space and compresses it such that the resulting word embeddings are good predictors for a target bilexical relation. In experiments we show that task-specific embeddings can benefit both the quality and efficiency in lexical prediction tasks.
Cite
@article{arxiv.1412.7004,
title = {Tailoring Word Embeddings for Bilexical Predictions: An Experimental Comparison},
author = {Pranava Swaroop Madhyastha and Xavier Carreras and Ariadna Quattoni},
journal= {arXiv preprint arXiv:1412.7004},
year = {2015}
}
Comments
Accepted as a workshop contribution at ICLR 2015