Semantic Relation Classification via Convolutional Neural Networks with Simple Negative Sampling.
K. Xu, Y. Feng, S. Huang, и D. Zhao. Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing EMNLP, стр. 536–540. (2015)cite arxiv:1506.07650.
Аннотация
Syntactic features play an essential role in identifying relationship in a
sentence. Previous neural network models often suffer from irrelevant
information introduced when subjects and objects are in a long distance. In
this paper, we propose to learn more robust relation representations from the
shortest dependency path through a convolution neural network. We further
propose a straightforward negative sampling strategy to improve the assignment
of subjects and objects. Experimental results show that our method outperforms
the state-of-the-art methods on the SemEval-2010 Task 8 dataset.
Описание
Semantic Relation Classification via Convolutional Neural Networks with
Simple Negative Sampling
%0 Conference Paper
%1 xu2015semantic
%A Xu, Kun
%A Feng, Yansong
%A Huang, Songfang
%A Zhao, Dongyan
%B Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing EMNLP
%D 2015
%K classification cnn networks neural relation sampling semantic semeval semeval10
%P 536–540
%T Semantic Relation Classification via Convolutional Neural Networks with Simple Negative Sampling.
%U http://dblp.uni-trier.de/db/conf/emnlp/emnlp2015.html#XuFHZ15
%X Syntactic features play an essential role in identifying relationship in a
sentence. Previous neural network models often suffer from irrelevant
information introduced when subjects and objects are in a long distance. In
this paper, we propose to learn more robust relation representations from the
shortest dependency path through a convolution neural network. We further
propose a straightforward negative sampling strategy to improve the assignment
of subjects and objects. Experimental results show that our method outperforms
the state-of-the-art methods on the SemEval-2010 Task 8 dataset.
@inproceedings{xu2015semantic,
abstract = {Syntactic features play an essential role in identifying relationship in a
sentence. Previous neural network models often suffer from irrelevant
information introduced when subjects and objects are in a long distance. In
this paper, we propose to learn more robust relation representations from the
shortest dependency path through a convolution neural network. We further
propose a straightforward negative sampling strategy to improve the assignment
of subjects and objects. Experimental results show that our method outperforms
the state-of-the-art methods on the SemEval-2010 Task 8 dataset.},
added-at = {2017-10-05T22:37:35.000+0200},
author = {Xu, Kun and Feng, Yansong and Huang, Songfang and Zhao, Dongyan},
biburl = {https://www.bibsonomy.org/bibtex/21240d1169e1ee92205dda5790d1fad99/schwemmlein},
booktitle = {Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing [EMNLP]},
description = {Semantic Relation Classification via Convolutional Neural Networks with
Simple Negative Sampling},
interhash = {934f718044adc29d9e293b6d2b7befcd},
intrahash = {1240d1169e1ee92205dda5790d1fad99},
keywords = {classification cnn networks neural relation sampling semantic semeval semeval10},
note = {cite arxiv:1506.07650},
pages = {536–540},
timestamp = {2017-10-06T09:33:40.000+0200},
title = {Semantic Relation Classification via Convolutional Neural Networks with Simple Negative Sampling.},
url = {http://dblp.uni-trier.de/db/conf/emnlp/emnlp2015.html#XuFHZ15},
year = 2015
}