Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction

Chen Lin; Timothy A. Miller; Dmitriy Dligach; Hadi Amiri; Steven Bethard; Guergana Savova

Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction

Chen Lin
, Timothy A. Miller
, Dmitriy Dligach
, Hadi Amiri
, Steven Bethard
, Guergana Savova

Information Science

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

36 Scopus citations

Abstract

Neural network models are oftentimes restricted by limited labeled instances and resort to advanced architectures and features for cutting edge performance. We propose to build a recurrent neural network with multiple semantically heterogeneous embeddings within a self-training framework. Our framework makes use of labeled, unlabeled, and social media data, operates on basic features, and is scalable and generalizable. With this method, we establish the state-of-the-art result for both in- and cross-domain for a clinical temporal relation extraction task.

Original language	English (US)
Title of host publication	EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop
Publisher	Association for Computational Linguistics (ACL)
Pages	165-176
Number of pages	12
ISBN (Electronic)	9781948087742
State	Published - 2018
Event	9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018, co-located with EMNLP 2018 - Brussels, Belgium Duration: Oct 31 2018 → …

Publication series

Name	EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop

Conference

Conference	9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018, co-located with EMNLP 2018
Country/Territory	Belgium
City	Brussels
Period	10/31/18 → …

ASJC Scopus subject areas

Information Systems
Computer Science Applications
Computational Theory and Mathematics

Cite this

Lin, C., Miller, T. A., Dligach, D., Amiri, H., Bethard, S., & Savova, G. (2018). Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction. In EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop (pp. 165-176). (EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop). Association for Computational Linguistics (ACL).

Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction. / Lin, Chen; Miller, Timothy A.; Dligach, Dmitriy et al.
EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop. Association for Computational Linguistics (ACL), 2018. p. 165-176 (EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop).

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

Lin, C, Miller, TA, Dligach, D, Amiri, H, Bethard, S & Savova, G 2018, Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction. in EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop. EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop, Association for Computational Linguistics (ACL), pp. 165-176, 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018, co-located with EMNLP 2018, Brussels, Belgium, 10/31/18.

Lin C, Miller TA, Dligach D, Amiri H, Bethard S, Savova G. Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction. In EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop. Association for Computational Linguistics (ACL). 2018. p. 165-176. (EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop).

Lin, Chen ; Miller, Timothy A. ; Dligach, Dmitriy et al. / Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction. EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop. Association for Computational Linguistics (ACL), 2018. pp. 165-176 (EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop).

@inproceedings{a6236bf0f26041dc9847502efd8fa688,

title = "Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction",

abstract = "Neural network models are oftentimes restricted by limited labeled instances and resort to advanced architectures and features for cutting edge performance. We propose to build a recurrent neural network with multiple semantically heterogeneous embeddings within a self-training framework. Our framework makes use of labeled, unlabeled, and social media data, operates on basic features, and is scalable and generalizable. With this method, we establish the state-of-the-art result for both in- and cross-domain for a clinical temporal relation extraction task.",

author = "Chen Lin and Miller, \{Timothy A.\} and Dmitriy Dligach and Hadi Amiri and Steven Bethard and Guergana Savova",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics.; 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018, co-located with EMNLP 2018 ; Conference date: 31-10-2018",

year = "2018",

language = "English (US)",

series = "EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop",

publisher = "Association for Computational Linguistics (ACL)",

pages = "165--176",

booktitle = "EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop",

}

TY - GEN

T1 - Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction

AU - Lin, Chen

AU - Miller, Timothy A.

AU - Dligach, Dmitriy

AU - Amiri, Hadi

AU - Bethard, Steven

AU - Savova, Guergana

PY - 2018

Y1 - 2018

N2 - Neural network models are oftentimes restricted by limited labeled instances and resort to advanced architectures and features for cutting edge performance. We propose to build a recurrent neural network with multiple semantically heterogeneous embeddings within a self-training framework. Our framework makes use of labeled, unlabeled, and social media data, operates on basic features, and is scalable and generalizable. With this method, we establish the state-of-the-art result for both in- and cross-domain for a clinical temporal relation extraction task.

AB - Neural network models are oftentimes restricted by limited labeled instances and resort to advanced architectures and features for cutting edge performance. We propose to build a recurrent neural network with multiple semantically heterogeneous embeddings within a self-training framework. Our framework makes use of labeled, unlabeled, and social media data, operates on basic features, and is scalable and generalizable. With this method, we establish the state-of-the-art result for both in- and cross-domain for a clinical temporal relation extraction task.

UR - https://www.scopus.com/pages/publications/85066805645

UR - https://www.scopus.com/pages/publications/85066805645#tab=citedBy

M3 - Conference contribution

AN - SCOPUS:85066805645

T3 - EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop

SP - 165

EP - 176

BT - EMNLP 2018 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018 - Proceedings of the Workshop

PB - Association for Computational Linguistics (ACL)

T2 - 9th International Workshop on Health Text Mining and Information Analysis, LOUHI 2018, co-located with EMNLP 2018

Y2 - 31 October 2018

ER -

Self-training improves Recurrent Neural Networks performance for Temporal Relation Extraction

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this