Joint embedding of hierarchical structure and context for entity disambiguation

Shuangshuang Cai, Mizuho Iwaihara

Research output: Contribution to journalArticle

Abstract

Entity linking refers to the task of constructing links between the mentions of context and the description pages from knowledge base. Due to the polysemy phenomenon, the key issue of entity linking is entity disambiguation. To simplify the goal of entity disambiguation, the main problem is choosing the correct entity from candidates. In this paper, we propose a novel embedding method specifically designed for entity disambiguation. Existing distributed representations are limited in utilizing structured knowledge from knowledge bases such as Wikipedia. Our method jointly maps the information from hierarchical structure of knowledge and context words. We extend the continuous bags-of-words model by adding hierarchical categories and hyperlink structure. So far, we have trained a joint model which adds category information. We demonstrate the utility of our proposed approach on an entity relatedness dataset. The results show that our jointly embedding model is superior to the model simply using context words. In addition, we do disambiguation experiment on a dataset, and the results show slight superiority of the novel embedding model.

Original languageEnglish
Pages (from-to)61-73
Number of pages13
JournalStudies in Computational Intelligence
Volume721
DOIs
Publication statusPublished - 2018

Fingerprint

Experiments

Keywords

  • Categorical hierarchy
  • Entity disambiguation
  • Hyperlinks
  • Semantic relatedness

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Joint embedding of hierarchical structure and context for entity disambiguation. / Cai, Shuangshuang; Iwaihara, Mizuho.

In: Studies in Computational Intelligence, Vol. 721, 2018, p. 61-73.

Research output: Contribution to journalArticle

@article{44b679a62cdd4d0fb105040c45ee5f9f,
title = "Joint embedding of hierarchical structure and context for entity disambiguation",
abstract = "Entity linking refers to the task of constructing links between the mentions of context and the description pages from knowledge base. Due to the polysemy phenomenon, the key issue of entity linking is entity disambiguation. To simplify the goal of entity disambiguation, the main problem is choosing the correct entity from candidates. In this paper, we propose a novel embedding method specifically designed for entity disambiguation. Existing distributed representations are limited in utilizing structured knowledge from knowledge bases such as Wikipedia. Our method jointly maps the information from hierarchical structure of knowledge and context words. We extend the continuous bags-of-words model by adding hierarchical categories and hyperlink structure. So far, we have trained a joint model which adds category information. We demonstrate the utility of our proposed approach on an entity relatedness dataset. The results show that our jointly embedding model is superior to the model simply using context words. In addition, we do disambiguation experiment on a dataset, and the results show slight superiority of the novel embedding model.",
keywords = "Categorical hierarchy, Entity disambiguation, Hyperlinks, Semantic relatedness",
author = "Shuangshuang Cai and Mizuho Iwaihara",
year = "2018",
doi = "10.1007/978-3-319-62048-0_5",
language = "English",
volume = "721",
pages = "61--73",
journal = "Studies in Computational Intelligence",
issn = "1860-949X",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Joint embedding of hierarchical structure and context for entity disambiguation

AU - Cai, Shuangshuang

AU - Iwaihara, Mizuho

PY - 2018

Y1 - 2018

N2 - Entity linking refers to the task of constructing links between the mentions of context and the description pages from knowledge base. Due to the polysemy phenomenon, the key issue of entity linking is entity disambiguation. To simplify the goal of entity disambiguation, the main problem is choosing the correct entity from candidates. In this paper, we propose a novel embedding method specifically designed for entity disambiguation. Existing distributed representations are limited in utilizing structured knowledge from knowledge bases such as Wikipedia. Our method jointly maps the information from hierarchical structure of knowledge and context words. We extend the continuous bags-of-words model by adding hierarchical categories and hyperlink structure. So far, we have trained a joint model which adds category information. We demonstrate the utility of our proposed approach on an entity relatedness dataset. The results show that our jointly embedding model is superior to the model simply using context words. In addition, we do disambiguation experiment on a dataset, and the results show slight superiority of the novel embedding model.

AB - Entity linking refers to the task of constructing links between the mentions of context and the description pages from knowledge base. Due to the polysemy phenomenon, the key issue of entity linking is entity disambiguation. To simplify the goal of entity disambiguation, the main problem is choosing the correct entity from candidates. In this paper, we propose a novel embedding method specifically designed for entity disambiguation. Existing distributed representations are limited in utilizing structured knowledge from knowledge bases such as Wikipedia. Our method jointly maps the information from hierarchical structure of knowledge and context words. We extend the continuous bags-of-words model by adding hierarchical categories and hyperlink structure. So far, we have trained a joint model which adds category information. We demonstrate the utility of our proposed approach on an entity relatedness dataset. The results show that our jointly embedding model is superior to the model simply using context words. In addition, we do disambiguation experiment on a dataset, and the results show slight superiority of the novel embedding model.

KW - Categorical hierarchy

KW - Entity disambiguation

KW - Hyperlinks

KW - Semantic relatedness

UR - http://www.scopus.com/inward/record.url?scp=85022025737&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85022025737&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-62048-0_5

DO - 10.1007/978-3-319-62048-0_5

M3 - Article

VL - 721

SP - 61

EP - 73

JO - Studies in Computational Intelligence

JF - Studies in Computational Intelligence

SN - 1860-949X

ER -