Modeling Complex Relationship Paths for Knowledge Graph Completion

Ping ZENG; Qingping TAN; Xiankai MENG; Haoyu ZHANG; Jianjun XU

doi:10.1587/transinf.2017EDP7398

Modeling Complex Relationship Paths for Knowledge Graph Completion

Ping ZENG, Qingping TAN, Xiankai MENG, Haoyu ZHANG, Jianjun XU

Full Text Views

0

Cite this

Summary :

Determining the validity of knowledge triples and filling in the missing entities or relationships in the knowledge graph are the crucial tasks for large-scale knowledge graph completion. So far, the main solutions use machine learning methods to learn the low-dimensional distributed representations of entities and relationships to complete the knowledge graph. Among them, translation models obtain excellent performance. However, the proposed translation models do not adequately consider the indirect relationships among entities, affecting the precision of the representation. Based on the long short-term memory neural network and existing translation models, we propose a multiple-module hybrid neural network model called TransP. By modeling the entity paths and their relationship paths, TransP can effectively excavate the indirect relationships among the entities, and thus, improve the quality of knowledge graph completion tasks. Experimental results show that TransP outperforms state-of-the-art models in the entity prediction task, and achieves the comparable performance with previous models in the relationship prediction task.

Publication: IEICE TRANSACTIONS on Information Vol.E101-D No.5 pp.1393-1400

Publication Date: 2018/05/01

Publicized: 2018/02/20

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2017EDP7398

Type of Manuscript: PAPER

Category: Artificial Intelligence, Data Mining

Authors

Ping ZENG
  National University of Defense Technology,National Key Laboratory for Parallel and Distributed Processing
Qingping TAN
  National University of Defense Technology,National Key Laboratory for Parallel and Distributed Processing
Xiankai MENG
  National University of Defense Technology,National Key Laboratory for Parallel and Distributed Processing
Haoyu ZHANG
  National University of Defense Technology,National Key Laboratory for Parallel and Distributed Processing
Jianjun XU
  National University of Defense Technology,National Key Laboratory for Parallel and Distributed Processing

Keyword

knowledge graph completion, knowledge representation learning, knowledge graph

Cite this

Copy

Ping ZENG, Qingping TAN, Xiankai MENG, Haoyu ZHANG, Jianjun XU, "Modeling Complex Relationship Paths for Knowledge Graph Completion" in IEICE TRANSACTIONS on Information, vol. E101-D, no. 5, pp. 1393-1400, May 2018, doi: 10.1587/transinf.2017EDP7398.
Abstract: Determining the validity of knowledge triples and filling in the missing entities or relationships in the knowledge graph are the crucial tasks for large-scale knowledge graph completion. So far, the main solutions use machine learning methods to learn the low-dimensional distributed representations of entities and relationships to complete the knowledge graph. Among them, translation models obtain excellent performance. However, the proposed translation models do not adequately consider the indirect relationships among entities, affecting the precision of the representation. Based on the long short-term memory neural network and existing translation models, we propose a multiple-module hybrid neural network model called TransP. By modeling the entity paths and their relationship paths, TransP can effectively excavate the indirect relationships among the entities, and thus, improve the quality of knowledge graph completion tasks. Experimental results show that TransP outperforms state-of-the-art models in the entity prediction task, and achieves the comparable performance with previous models in the relationship prediction task.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017EDP7398/_p

Copy

@ARTICLE{e101-d_5_1393,
author={Ping ZENG, Qingping TAN, Xiankai MENG, Haoyu ZHANG, Jianjun XU, },
journal={IEICE TRANSACTIONS on Information},
title={Modeling Complex Relationship Paths for Knowledge Graph Completion},
year={2018},
volume={E101-D},
number={5},
pages={1393-1400},
abstract={Determining the validity of knowledge triples and filling in the missing entities or relationships in the knowledge graph are the crucial tasks for large-scale knowledge graph completion. So far, the main solutions use machine learning methods to learn the low-dimensional distributed representations of entities and relationships to complete the knowledge graph. Among them, translation models obtain excellent performance. However, the proposed translation models do not adequately consider the indirect relationships among entities, affecting the precision of the representation. Based on the long short-term memory neural network and existing translation models, we propose a multiple-module hybrid neural network model called TransP. By modeling the entity paths and their relationship paths, TransP can effectively excavate the indirect relationships among the entities, and thus, improve the quality of knowledge graph completion tasks. Experimental results show that TransP outperforms state-of-the-art models in the entity prediction task, and achieves the comparable performance with previous models in the relationship prediction task.},
keywords={},
doi={10.1587/transinf.2017EDP7398},
ISSN={1745-1361},
month={May},}

Copy

TY - JOUR
TI - Modeling Complex Relationship Paths for Knowledge Graph Completion
T2 - IEICE TRANSACTIONS on Information
SP - 1393
EP - 1400
AU - Ping ZENG
AU - Qingping TAN
AU - Xiankai MENG
AU - Haoyu ZHANG
AU - Jianjun XU
PY - 2018
DO - 10.1587/transinf.2017EDP7398
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2018
AB - Determining the validity of knowledge triples and filling in the missing entities or relationships in the knowledge graph are the crucial tasks for large-scale knowledge graph completion. So far, the main solutions use machine learning methods to learn the low-dimensional distributed representations of entities and relationships to complete the knowledge graph. Among them, translation models obtain excellent performance. However, the proposed translation models do not adequately consider the indirect relationships among entities, affecting the precision of the representation. Based on the long short-term memory neural network and existing translation models, we propose a multiple-module hybrid neural network model called TransP. By modeling the entity paths and their relationship paths, TransP can effectively excavate the indirect relationships among the entities, and thus, improve the quality of knowledge graph completion tasks. Experimental results show that TransP outperforms state-of-the-art models in the entity prediction task, and achieves the comparable performance with previous models in the relationship prediction task.
ER -