Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations

Tharindu Ranasinghe; Constantin Orasan; Ruslan Mitkov

doi:10.26615/978-954-452-056-4_115

Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations

Tharindu Ranasinghe, Constantin Orasan, Ruslan Mitkov

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

Calculating Semantic Textual Similarity (STS) plays a significant role in many applications such as question answering, document summarisation, information retrieval and information extraction. All modern state of the art STS methods rely on word embeddings one way or another. The recently introduced contextualised word embeddings have proved more effective than standard word embeddings in many natural language processing tasks. This paper evaluates the impact of several contextualised word embeddings on unsupervised STS methods and compares it with the existing supervised/unsupervised STS methods for different datasets in different languages and different domains

Original language	English
Title of host publication	Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)
Pages	994–1003
DOIs	https://doi.org/10.26615/978-954-452-056-4_115
Publication status	Published - Sept 2019
Event	Recent Advances in Natural Language Processing - Varna, Bulgaria Duration: 2 Sept 2019 → 4 Sept 2019 https://www.ranlp.org/archive/ranlp2019/start.php

Conference

Conference	Recent Advances in Natural Language Processing
Abbreviated title	RANLP 2019
Country/Territory	Bulgaria
City	Varna
Period	2/09/19 → 4/09/19
Internet address	https://www.ranlp.org/archive/ranlp2019/start.php

Access to Document

10.26615/978-954-452-056-4_115

Cite this

@inproceedings{a80a1618fd4b4741848ed94e8f5b9eb9,

title = "Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations",

abstract = "Calculating Semantic Textual Similarity (STS) plays a significant role in many applications such as question answering, document summarisation, information retrieval and information extraction. All modern state of the art STS methods rely on word embeddings one way or another. The recently introduced contextualised word embeddings have proved more effective than standard word embeddings in many natural language processing tasks. This paper evaluates the impact of several contextualised word embeddings on unsupervised STS methods and compares it with the existing supervised/unsupervised STS methods for different datasets in different languages and different domains",

author = "Tharindu Ranasinghe and Constantin Orasan and Ruslan Mitkov",

year = "2019",

month = sep,

doi = "10.26615/978-954-452-056-4_115",

language = "English",

pages = "994–1003",

booktitle = "Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)",

note = "Recent Advances in Natural Language Processing, RANLP 2019 ; Conference date: 02-09-2019 Through 04-09-2019",

url = "https://www.ranlp.org/archive/ranlp2019/start.php",

}

Ranasinghe, T, Orasan, C & Mitkov, R 2019, Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations. in Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019). pp. 994–1003, Recent Advances in Natural Language Processing, Varna, Bulgaria, 2/09/19. https://doi.org/10.26615/978-954-452-056-4_115

Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations. / Ranasinghe, Tharindu; Orasan, Constantin; Mitkov, Ruslan.
Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019). 2019. p. 994–1003.

Research output: Chapter in Book/Published conference output › Conference publication

TY - GEN

T1 - Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations

AU - Ranasinghe, Tharindu

AU - Orasan, Constantin

AU - Mitkov, Ruslan

PY - 2019/9

Y1 - 2019/9

N2 - Calculating Semantic Textual Similarity (STS) plays a significant role in many applications such as question answering, document summarisation, information retrieval and information extraction. All modern state of the art STS methods rely on word embeddings one way or another. The recently introduced contextualised word embeddings have proved more effective than standard word embeddings in many natural language processing tasks. This paper evaluates the impact of several contextualised word embeddings on unsupervised STS methods and compares it with the existing supervised/unsupervised STS methods for different datasets in different languages and different domains

AB - Calculating Semantic Textual Similarity (STS) plays a significant role in many applications such as question answering, document summarisation, information retrieval and information extraction. All modern state of the art STS methods rely on word embeddings one way or another. The recently introduced contextualised word embeddings have proved more effective than standard word embeddings in many natural language processing tasks. This paper evaluates the impact of several contextualised word embeddings on unsupervised STS methods and compares it with the existing supervised/unsupervised STS methods for different datasets in different languages and different domains

UR - https://www.aclweb.org/anthology/R19-1115

U2 - 10.26615/978-954-452-056-4_115

DO - 10.26615/978-954-452-056-4_115

M3 - Conference publication

SP - 994

EP - 1003

BT - Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)

T2 - Recent Advances in Natural Language Processing

Y2 - 2 September 2019 through 4 September 2019

ER -

Enhancing Unsupervised Sentence Similarity Methods with Deep Contextualised Word Representations

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this