Using natural language to integrate, evaluate, and optimize extracted knowledge bases

Doug Downey; Chandra Sekhar Bhagavatula; Alexander Yates

doi:10.1145/2509558.2509569

Using natural language to integrate, evaluate, and optimize extracted knowledge bases

Doug Downey, Chandra Sekhar Bhagavatula, Alexander Yates

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

Web Information Extraction (WIE) systems extract billions of unique facts, but integrating the assertions into a coherent knowledge base and evaluating across different WIE techniques remains a challenge. We propose a framework that utilizes natural language to integrate and evaluate extracted knowledge bases (KBs). In the framework, KBs are integrated by exchanging probability distributions over natural language, and evaluated by how well the output distributions predict held-out text. We describe the advantages of the approach, and detail remaining research challenges.

Original language	English (US)
Title of host publication	AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013
Pages	61-66
Number of pages	6
DOIs	https://doi.org/10.1145/2509558.2509569
State	Published - 2013
Event	2013 Workshop on Automated Knowledge Base Construction, AKBC 2013 - Co-located with CIKM 2013 - San Francisco, CA, United States Duration: Oct 27 2013 → Oct 28 2013

Publication series

Name	AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013

Other

Other	2013 Workshop on Automated Knowledge Base Construction, AKBC 2013 - Co-located with CIKM 2013
Country/Territory	United States
City	San Francisco, CA
Period	10/27/13 → 10/28/13

Keywords

knowledge extraction
knowledge integration
language modeling

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications

Access to Document

10.1145/2509558.2509569

Cite this

Downey, D., Bhagavatula, C. S., & Yates, A. (2013). Using natural language to integrate, evaluate, and optimize extracted knowledge bases. In AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013 (pp. 61-66). (AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013). https://doi.org/10.1145/2509558.2509569

Downey, Doug ; Bhagavatula, Chandra Sekhar ; Yates, Alexander. / Using natural language to integrate, evaluate, and optimize extracted knowledge bases. AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013. 2013. pp. 61-66 (AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013).

@inproceedings{717889975ef645e8afebb2e962a81730,

title = "Using natural language to integrate, evaluate, and optimize extracted knowledge bases",

abstract = "Web Information Extraction (WIE) systems extract billions of unique facts, but integrating the assertions into a coherent knowledge base and evaluating across different WIE techniques remains a challenge. We propose a framework that utilizes natural language to integrate and evaluate extracted knowledge bases (KBs). In the framework, KBs are integrated by exchanging probability distributions over natural language, and evaluated by how well the output distributions predict held-out text. We describe the advantages of the approach, and detail remaining research challenges.",

keywords = "knowledge extraction, knowledge integration, language modeling",

author = "Doug Downey and Bhagavatula, {Chandra Sekhar} and Alexander Yates",

year = "2013",

doi = "10.1145/2509558.2509569",

language = "English (US)",

isbn = "9781450324113",

series = "AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013",

pages = "61--66",

booktitle = "AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013",

note = "2013 Workshop on Automated Knowledge Base Construction, AKBC 2013 - Co-located with CIKM 2013 ; Conference date: 27-10-2013 Through 28-10-2013",

}

Downey, D, Bhagavatula, CS & Yates, A 2013, Using natural language to integrate, evaluate, and optimize extracted knowledge bases. in AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013. AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013, pp. 61-66, 2013 Workshop on Automated Knowledge Base Construction, AKBC 2013 - Co-located with CIKM 2013, San Francisco, CA, United States, 10/27/13. https://doi.org/10.1145/2509558.2509569

Using natural language to integrate, evaluate, and optimize extracted knowledge bases. / Downey, Doug; Bhagavatula, Chandra Sekhar; Yates, Alexander.
AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013. 2013. p. 61-66 (AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Using natural language to integrate, evaluate, and optimize extracted knowledge bases

AU - Downey, Doug

AU - Bhagavatula, Chandra Sekhar

AU - Yates, Alexander

PY - 2013

Y1 - 2013

N2 - Web Information Extraction (WIE) systems extract billions of unique facts, but integrating the assertions into a coherent knowledge base and evaluating across different WIE techniques remains a challenge. We propose a framework that utilizes natural language to integrate and evaluate extracted knowledge bases (KBs). In the framework, KBs are integrated by exchanging probability distributions over natural language, and evaluated by how well the output distributions predict held-out text. We describe the advantages of the approach, and detail remaining research challenges.

AB - Web Information Extraction (WIE) systems extract billions of unique facts, but integrating the assertions into a coherent knowledge base and evaluating across different WIE techniques remains a challenge. We propose a framework that utilizes natural language to integrate and evaluate extracted knowledge bases (KBs). In the framework, KBs are integrated by exchanging probability distributions over natural language, and evaluated by how well the output distributions predict held-out text. We describe the advantages of the approach, and detail remaining research challenges.

KW - knowledge extraction

KW - knowledge integration

KW - language modeling

UR - http://www.scopus.com/inward/record.url?scp=84888162528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888162528&partnerID=8YFLogxK

U2 - 10.1145/2509558.2509569

DO - 10.1145/2509558.2509569

M3 - Conference contribution

AN - SCOPUS:84888162528

SN - 9781450324113

T3 - AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013

SP - 61

EP - 66

BT - AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013

T2 - 2013 Workshop on Automated Knowledge Base Construction, AKBC 2013 - Co-located with CIKM 2013

Y2 - 27 October 2013 through 28 October 2013

ER -

Downey D, Bhagavatula CS, Yates A. Using natural language to integrate, evaluate, and optimize extracted knowledge bases. In AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013. 2013. p. 61-66. (AKBC 2013 - Proceedings of the 2013 Workshop on Automated Knowledge Base Construction, Co-located with CIKM 2013). doi: 10.1145/2509558.2509569

Using natural language to integrate, evaluate, and optimize extracted knowledge bases

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this