Using entity information from a knowledge base to improve relation extraction

Lan Du; Anish Kumar; Mark Johnson; Massimiliano Ciaramita

Using entity information from a knowledge base to improve relation extraction

Lan Du, Anish Kumar, Mark Johnson, Massimiliano Ciaramita

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

Abstract

Relation extraction is the task of extracting predicate-argument relationships
between entities from natural language text. This paper investigates whether background information about entities available in knowledge bases such as FreeBase can be used to improve the accuracy of a state-of-the-art relation extraction system. We describe a simple and effective way of incorporating FreeBase’s notable types into a state-of-the-art relation extraction system (Riedel et al., 2013). Experimental results show that our notable typebased
system achieves an average 7.5% weighted MAP score improvement. To
understand where the notable type information contributes the most, we perform a series of ablation experiments. Results show that the notable type information improves relation extraction more than NER labels alone across a wide range of entity types and relations.

Original language	English
Title of host publication	Australasian Language Technology Association Workshop 2015
Subtitle of host publication	Proceedings of the Workshop
Editors	Ben Hachey, Kellie Webster
Place of Publication	Melbourne Vic Australia
Publisher	Australasian Language Technology Association
Pages	31-38
Number of pages	8
Publication status	Published - 2015
Event	Australasian Language Technology Association Workshop 2015 - University of Western Sydney, Parramatta, Australia Duration: 8 Dec 2015 → 9 Dec 2015 Conference number: 13th https://www.aclweb.org/anthology/events/alta-2015/ (Proceedings)

Publication series

Name	Proceedings of the Australasian Language Technology Workshop
Publisher	Australasian Language Technology Association
Volume	13
ISSN (Print)	1834-7037

Conference

Conference	Australasian Language Technology Association Workshop 2015
Abbreviated title	ALTAW 2015
Country/Territory	Australia
City	Parramatta
Period	8/12/15 → 9/12/15
Other	ALTA 2015 will be co-located with the 2nd Conference of the Australian Music and Psychology Society (AMPS 2015), the 20th Australasian Document Computing Symposium 2014 (ADCS 2015) and the 46th Australian Linguistic Society conference (ALS 2015).
Internet address	https://www.aclweb.org/anthology/events/alta-2015/ (Proceedings)

Access to Document

240162824-oaFinal published version, 1.03 MB

http://www.aclweb.org/anthology/U15-1004

Cite this

Du, L., Kumar, A., Johnson, M., & Ciaramita, M. (2015). Using entity information from a knowledge base to improve relation extraction. In B. Hachey, & K. Webster (Eds.), Australasian Language Technology Association Workshop 2015: Proceedings of the Workshop (pp. 31-38). (Proceedings of the Australasian Language Technology Workshop; Vol. 13). Australasian Language Technology Association. http://www.aclweb.org/anthology/U15-1004

Du, Lan ; Kumar, Anish ; Johnson, Mark et al. / Using entity information from a knowledge base to improve relation extraction. Australasian Language Technology Association Workshop 2015: Proceedings of the Workshop. editor / Ben Hachey ; Kellie Webster. Melbourne Vic Australia : Australasian Language Technology Association, 2015. pp. 31-38 (Proceedings of the Australasian Language Technology Workshop).

@inproceedings{47fd35af8e6348d79152c2da69a95dba,

title = "Using entity information from a knowledge base to improve relation extraction",

abstract = "Relation extraction is the task of extracting predicate-argument relationshipsbetween entities from natural language text. This paper investigates whether background information about entities available in knowledge bases such as FreeBase can be used to improve the accuracy of a state-of-the-art relation extraction system. We describe a simple and effective way of incorporating FreeBase{\textquoteright}s notable types into a state-of-the-art relation extraction system (Riedel et al., 2013). Experimental results show that our notable typebasedsystem achieves an average 7.5% weighted MAP score improvement. Tounderstand where the notable type information contributes the most, we perform a series of ablation experiments. Results show that the notable type information improves relation extraction more than NER labels alone across a wide range of entity types and relations.",

author = "Lan Du and Anish Kumar and Mark Johnson and Massimiliano Ciaramita",

year = "2015",

language = "English",

series = "Proceedings of the Australasian Language Technology Workshop",

publisher = "Australasian Language Technology Association",

pages = "31--38",

editor = "Ben Hachey and Kellie Webster",

booktitle = "Australasian Language Technology Association Workshop 2015",

note = "Australasian Language Technology Association Workshop 2015, ALTAW 2015 ; Conference date: 08-12-2015 Through 09-12-2015",

url = "https://www.aclweb.org/anthology/events/alta-2015/",

}

Du, L, Kumar, A, Johnson, M & Ciaramita, M 2015, Using entity information from a knowledge base to improve relation extraction. in B Hachey & K Webster (eds), Australasian Language Technology Association Workshop 2015: Proceedings of the Workshop. Proceedings of the Australasian Language Technology Workshop, vol. 13, Australasian Language Technology Association, Melbourne Vic Australia, pp. 31-38, Australasian Language Technology Association Workshop 2015, Parramatta, New South Wales, Australia, 8/12/15. <http://www.aclweb.org/anthology/U15-1004>

Using entity information from a knowledge base to improve relation extraction. / Du, Lan; Kumar, Anish; Johnson, Mark et al.
Australasian Language Technology Association Workshop 2015: Proceedings of the Workshop. ed. / Ben Hachey; Kellie Webster. Melbourne Vic Australia: Australasian Language Technology Association, 2015. p. 31-38 (Proceedings of the Australasian Language Technology Workshop; Vol. 13).

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

TY - GEN

T1 - Using entity information from a knowledge base to improve relation extraction

AU - Du, Lan

AU - Kumar, Anish

AU - Johnson, Mark

AU - Ciaramita, Massimiliano

N1 - Conference code: 13th

PY - 2015

Y1 - 2015

N2 - Relation extraction is the task of extracting predicate-argument relationshipsbetween entities from natural language text. This paper investigates whether background information about entities available in knowledge bases such as FreeBase can be used to improve the accuracy of a state-of-the-art relation extraction system. We describe a simple and effective way of incorporating FreeBase’s notable types into a state-of-the-art relation extraction system (Riedel et al., 2013). Experimental results show that our notable typebasedsystem achieves an average 7.5% weighted MAP score improvement. Tounderstand where the notable type information contributes the most, we perform a series of ablation experiments. Results show that the notable type information improves relation extraction more than NER labels alone across a wide range of entity types and relations.

AB - Relation extraction is the task of extracting predicate-argument relationshipsbetween entities from natural language text. This paper investigates whether background information about entities available in knowledge bases such as FreeBase can be used to improve the accuracy of a state-of-the-art relation extraction system. We describe a simple and effective way of incorporating FreeBase’s notable types into a state-of-the-art relation extraction system (Riedel et al., 2013). Experimental results show that our notable typebasedsystem achieves an average 7.5% weighted MAP score improvement. Tounderstand where the notable type information contributes the most, we perform a series of ablation experiments. Results show that the notable type information improves relation extraction more than NER labels alone across a wide range of entity types and relations.

M3 - Conference Paper

T3 - Proceedings of the Australasian Language Technology Workshop

SP - 31

EP - 38

BT - Australasian Language Technology Association Workshop 2015

A2 - Hachey, Ben

A2 - Webster, Kellie

PB - Australasian Language Technology Association

CY - Melbourne Vic Australia

T2 - Australasian Language Technology Association Workshop 2015

Y2 - 8 December 2015 through 9 December 2015

ER -

Du L, Kumar A, Johnson M, Ciaramita M. Using entity information from a knowledge base to improve relation extraction. In Hachey B, Webster K, editors, Australasian Language Technology Association Workshop 2015: Proceedings of the Workshop. Melbourne Vic Australia: Australasian Language Technology Association. 2015. p. 31-38. (Proceedings of the Australasian Language Technology Workshop).