End-to-End Relation Extraction using LSTMs on Sequences and Tree
Structures
M. Miwa, and M. Bansal. (2016)cite arxiv:1601.00770Comment: Accepted for publication at the Association for Computational Linguistics (ACL), 2016. 13 pages, 1 figure, 6 tables.
Abstract
We present a novel end-to-end neural model to extract entities and relations
between them. Our recurrent neural network based model captures both word
sequence and dependency tree substructure information by stacking bidirectional
tree-structured LSTM-RNNs on bidirectional sequential LSTM-RNNs. This allows
our model to jointly represent both entities and relations with shared
parameters in a single model. We further encourage detection of entities during
training and use of entity information in relation extraction via entity
pretraining and scheduled sampling. Our model improves over the
state-of-the-art feature-based model on end-to-end relation extraction,
achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and
ACE2004, respectively. We also show that our LSTM-RNN based model compares
favorably to the state-of-the-art CNN based model (in F1-score) on nominal
relation classification (SemEval-2010 Task 8). Finally, we present an extensive
ablation analysis of several model components.
%0 Generic
%1 miwa2016endtoend
%A Miwa, Makoto
%A Bansal, Mohit
%D 2016
%K deep_learning lstm relex rnn
%T End-to-End Relation Extraction using LSTMs on Sequences and Tree
Structures
%U http://arxiv.org/abs/1601.00770
%X We present a novel end-to-end neural model to extract entities and relations
between them. Our recurrent neural network based model captures both word
sequence and dependency tree substructure information by stacking bidirectional
tree-structured LSTM-RNNs on bidirectional sequential LSTM-RNNs. This allows
our model to jointly represent both entities and relations with shared
parameters in a single model. We further encourage detection of entities during
training and use of entity information in relation extraction via entity
pretraining and scheduled sampling. Our model improves over the
state-of-the-art feature-based model on end-to-end relation extraction,
achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and
ACE2004, respectively. We also show that our LSTM-RNN based model compares
favorably to the state-of-the-art CNN based model (in F1-score) on nominal
relation classification (SemEval-2010 Task 8). Finally, we present an extensive
ablation analysis of several model components.
@misc{miwa2016endtoend,
abstract = {We present a novel end-to-end neural model to extract entities and relations
between them. Our recurrent neural network based model captures both word
sequence and dependency tree substructure information by stacking bidirectional
tree-structured LSTM-RNNs on bidirectional sequential LSTM-RNNs. This allows
our model to jointly represent both entities and relations with shared
parameters in a single model. We further encourage detection of entities during
training and use of entity information in relation extraction via entity
pretraining and scheduled sampling. Our model improves over the
state-of-the-art feature-based model on end-to-end relation extraction,
achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and
ACE2004, respectively. We also show that our LSTM-RNN based model compares
favorably to the state-of-the-art CNN based model (in F1-score) on nominal
relation classification (SemEval-2010 Task 8). Finally, we present an extensive
ablation analysis of several model components.},
added-at = {2018-02-22T11:28:00.000+0100},
author = {Miwa, Makoto and Bansal, Mohit},
biburl = {https://www.bibsonomy.org/bibtex/29b4a0cdc94f96521535d1901cb136800/dallmann},
description = {1601.00770.pdf},
interhash = {a31c3fa7e26048ce288d8af72c2dc6c3},
intrahash = {9b4a0cdc94f96521535d1901cb136800},
keywords = {deep_learning lstm relex rnn},
note = {cite arxiv:1601.00770Comment: Accepted for publication at the Association for Computational Linguistics (ACL), 2016. 13 pages, 1 figure, 6 tables},
timestamp = {2018-02-22T11:28:00.000+0100},
title = {End-to-End Relation Extraction using LSTMs on Sequences and Tree
Structures},
url = {http://arxiv.org/abs/1601.00770},
year = 2016
}