Contextualized Word Embeddings in a Neural Open Information Extraction Model

I. Sarhan; M. Spruit

doi:10.1007/978-3-030-23281-8_31

Contextualized Word Embeddings in a Neural Open Information Extraction Model

I. Sarhan, M. Spruit

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Open Information Extraction (OIE) is a challenging task of extracting relation tuples from an unstructured corpus. While several OIE algorithms have been developed in the past decade, only few employ deep learning techniques. In this paper, a novel OIE neural model that leverages Recurrent Neural Networks (RNN) using Gated Recurrent Units (GRUs) is presented. Moreover, we integrate the innovative contextual word embeddings into our OIE model, which further enhances the performance. The results demonstrate that our proposed neural OIE model outperforms the existing state-of-art on two datasets.

Original language	English
Title of host publication	NLDB 2019: International Conference on Applications of Natural Language to Information Systems
Editors	E. et al. Métais
Publisher	Springer
Pages	359–367
Number of pages	9
Volume	11608
ISBN (Electronic)	978-3-030-23281-8
ISBN (Print)	978-3-030-23280-1
DOIs	https://doi.org/10.1007/978-3-030-23281-8_31
Publication status	Published - 2019

Keywords

Open Information Extraction
Word embeddings
RNN
GRU
LSTM

Access to Document

10.1007/978-3-030-23281-8_31

Sarhan-Spruit2019_Chapter_ContextualizedWordEmbeddingsInFinal published version, 516 KBLicence: Taverne

Cite this

@inproceedings{7515314b36e342c7b18e2348cb8f54ea,

title = "Contextualized Word Embeddings in a Neural Open Information Extraction Model",

abstract = "Open Information Extraction (OIE) is a challenging task of extracting relation tuples from an unstructured corpus. While several OIE algorithms have been developed in the past decade, only few employ deep learning techniques. In this paper, a novel OIE neural model that leverages Recurrent Neural Networks (RNN) using Gated Recurrent Units (GRUs) is presented. Moreover, we integrate the innovative contextual word embeddings into our OIE model, which further enhances the performance. The results demonstrate that our proposed neural OIE model outperforms the existing state-of-art on two datasets.",

keywords = "Open Information Extraction, Word embeddings, RNN, GRU, LSTM",

author = "I. Sarhan and M. Spruit",

year = "2019",

doi = "10.1007/978-3-030-23281-8_31",

language = "English",

isbn = "978-3-030-23280-1",

volume = "11608",

pages = "359–367",

editor = "M{\'e}tais, {E. et al.}",

booktitle = "NLDB 2019: International Conference on Applications of Natural Language to Information Systems",

publisher = "Springer",

address = "Germany",

}

Contextualized Word Embeddings in a Neural Open Information Extraction Model. / Sarhan, I.; Spruit, M.
NLDB 2019: International Conference on Applications of Natural Language to Information Systems. ed. / E. et al. Métais. Vol. 11608 Springer, 2019. p. 359–367.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Contextualized Word Embeddings in a Neural Open Information Extraction Model

AU - Sarhan, I.

AU - Spruit, M.

PY - 2019

Y1 - 2019

N2 - Open Information Extraction (OIE) is a challenging task of extracting relation tuples from an unstructured corpus. While several OIE algorithms have been developed in the past decade, only few employ deep learning techniques. In this paper, a novel OIE neural model that leverages Recurrent Neural Networks (RNN) using Gated Recurrent Units (GRUs) is presented. Moreover, we integrate the innovative contextual word embeddings into our OIE model, which further enhances the performance. The results demonstrate that our proposed neural OIE model outperforms the existing state-of-art on two datasets.

AB - Open Information Extraction (OIE) is a challenging task of extracting relation tuples from an unstructured corpus. While several OIE algorithms have been developed in the past decade, only few employ deep learning techniques. In this paper, a novel OIE neural model that leverages Recurrent Neural Networks (RNN) using Gated Recurrent Units (GRUs) is presented. Moreover, we integrate the innovative contextual word embeddings into our OIE model, which further enhances the performance. The results demonstrate that our proposed neural OIE model outperforms the existing state-of-art on two datasets.

KW - Open Information Extraction

KW - Word embeddings

KW - RNN

KW - GRU

KW - LSTM

U2 - 10.1007/978-3-030-23281-8_31

DO - 10.1007/978-3-030-23281-8_31

M3 - Conference contribution

SN - 978-3-030-23280-1

VL - 11608

SP - 359

EP - 367

BT - NLDB 2019: International Conference on Applications of Natural Language to Information Systems

A2 - Métais, E. et al.

PB - Springer

ER -

Contextualized Word Embeddings in a Neural Open Information Extraction Model

Abstract

Keywords

Access to Document

Fingerprint

Cite this