Active learning for relation type extension with local and global data views

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Relation extraction is the process of identifying instances of specified types of semantic relations in text; relation type extension involves extending a relation extraction system to recognize a new type of relation. We present LGCo-Testing, an active learning system for relation type extension based on local and global views of relation instances. Locally, we extract features from the sentence that contains the instance. Globally, we measure the distributional similarity between instances from a 2 billion token corpus. Evaluation on the ACE 2004 corpus shows that LGCo-Testing can reduce annotation cost by 97% while maintaining the performance level of supervised learning.

Original languageEnglish (US)
Title of host publicationCIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management
Pages1105-1112
Number of pages8
DOIs
StatePublished - 2012
Event21st ACM International Conference on Information and Knowledge Management, CIKM 2012 - Maui, HI, United States
Duration: Oct 29 2012Nov 2 2012

Other

Other21st ACM International Conference on Information and Knowledge Management, CIKM 2012
CountryUnited States
CityMaui, HI
Period10/29/1211/2/12

Fingerprint

Supervised learning
Testing
Learning systems
Semantics
Costs
Problem-Based Learning

Keywords

  • co-testing
  • co-training
  • distributional similarity
  • infactive learning
  • information extraction
  • inrelation extraction

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Sun, A., & Grishman, R. (2012). Active learning for relation type extension with local and global data views. In CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management (pp. 1105-1112) https://doi.org/10.1145/2396761.2398409

Active learning for relation type extension with local and global data views. / Sun, Ang; Grishman, Ralph.

CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management. 2012. p. 1105-1112.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Sun, A & Grishman, R 2012, Active learning for relation type extension with local and global data views. in CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management. pp. 1105-1112, 21st ACM International Conference on Information and Knowledge Management, CIKM 2012, Maui, HI, United States, 10/29/12. https://doi.org/10.1145/2396761.2398409
Sun A, Grishman R. Active learning for relation type extension with local and global data views. In CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management. 2012. p. 1105-1112 https://doi.org/10.1145/2396761.2398409
Sun, Ang ; Grishman, Ralph. / Active learning for relation type extension with local and global data views. CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management. 2012. pp. 1105-1112
@inproceedings{0a22066f5fc2406e8d6941a1b19352a4,
title = "Active learning for relation type extension with local and global data views",
abstract = "Relation extraction is the process of identifying instances of specified types of semantic relations in text; relation type extension involves extending a relation extraction system to recognize a new type of relation. We present LGCo-Testing, an active learning system for relation type extension based on local and global views of relation instances. Locally, we extract features from the sentence that contains the instance. Globally, we measure the distributional similarity between instances from a 2 billion token corpus. Evaluation on the ACE 2004 corpus shows that LGCo-Testing can reduce annotation cost by 97{\%} while maintaining the performance level of supervised learning.",
keywords = "co-testing, co-training, distributional similarity, infactive learning, information extraction, inrelation extraction",
author = "Ang Sun and Ralph Grishman",
year = "2012",
doi = "10.1145/2396761.2398409",
language = "English (US)",
isbn = "9781450311564",
pages = "1105--1112",
booktitle = "CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management",

}

TY - GEN

T1 - Active learning for relation type extension with local and global data views

AU - Sun, Ang

AU - Grishman, Ralph

PY - 2012

Y1 - 2012

N2 - Relation extraction is the process of identifying instances of specified types of semantic relations in text; relation type extension involves extending a relation extraction system to recognize a new type of relation. We present LGCo-Testing, an active learning system for relation type extension based on local and global views of relation instances. Locally, we extract features from the sentence that contains the instance. Globally, we measure the distributional similarity between instances from a 2 billion token corpus. Evaluation on the ACE 2004 corpus shows that LGCo-Testing can reduce annotation cost by 97% while maintaining the performance level of supervised learning.

AB - Relation extraction is the process of identifying instances of specified types of semantic relations in text; relation type extension involves extending a relation extraction system to recognize a new type of relation. We present LGCo-Testing, an active learning system for relation type extension based on local and global views of relation instances. Locally, we extract features from the sentence that contains the instance. Globally, we measure the distributional similarity between instances from a 2 billion token corpus. Evaluation on the ACE 2004 corpus shows that LGCo-Testing can reduce annotation cost by 97% while maintaining the performance level of supervised learning.

KW - co-testing

KW - co-training

KW - distributional similarity

KW - infactive learning

KW - information extraction

KW - inrelation extraction

UR - http://www.scopus.com/inward/record.url?scp=84871102140&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871102140&partnerID=8YFLogxK

U2 - 10.1145/2396761.2398409

DO - 10.1145/2396761.2398409

M3 - Conference contribution

SN - 9781450311564

SP - 1105

EP - 1112

BT - CIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management

ER -