Covering a sentence in form and meaning with fewer retrieved sentences

Yuan Liu; Yves Lepage

Covering a sentence in form and meaning with fewer retrieved sentences

Yuan Liu, Yves Lepage

Graduate School of Information, Production and Systems

Research output: Contribution to conference › Paper › peer-review

1 Citation (Scopus)

Abstract

Retrieving similar sentences from a given collection of sentences is essential in a range of applications. In this work, we propose a novel method to retrieve several sentences that cover an input sentence in form and meaning with minimal redundancy, so as to enhance the overall coverage quality of the output sentences. We focus on the hierarchical granularity levels of sentence pieces, matching from common or similar n-grams to finer-grained words o subwords, using techniques from similar sentence retrieval and monolingual phrase alignment. Our method shows promising source and target coverage evaluation results when applied to parallel corpora. This shows the potential of our approach if integrated into an example-based machine translation system.

Original language	English
Pages	436-445
Number of pages	10
Publication status	Published - 2021
Event	35th Pacific Asia Conference on Language, Information and Computation, PACLIC 2021 - Shanghai, China Duration: 2021 Nov 5 → 2021 Nov 7

Conference

Conference	35th Pacific Asia Conference on Language, Information and Computation, PACLIC 2021
Country/Territory	China
City	Shanghai
Period	21/11/5 → 21/11/7

ASJC Scopus subject areas

Artificial Intelligence
Human-Computer Interaction
Linguistics and Language

Cite this

@conference{b6b2ca9086054b9eac252915adb08c61,

title = "Covering a sentence in form and meaning with fewer retrieved sentences",

abstract = "Retrieving similar sentences from a given collection of sentences is essential in a range of applications. In this work, we propose a novel method to retrieve several sentences that cover an input sentence in form and meaning with minimal redundancy, so as to enhance the overall coverage quality of the output sentences. We focus on the hierarchical granularity levels of sentence pieces, matching from common or similar n-grams to finer-grained words o subwords, using techniques from similar sentence retrieval and monolingual phrase alignment. Our method shows promising source and target coverage evaluation results when applied to parallel corpora. This shows the potential of our approach if integrated into an example-based machine translation system.",

author = "Yuan Liu and Yves Lepage",

note = "Funding Information: This research was supported in part by two grants-in-aid from the Japanese Society for the Promotion of Science (JSPS): no 18K11447 entitled “Self-explainable and fast-to-train example-based machine translation using neural networks” and no 18K11446 entitled “Natural language processing for academic writing in English” . Publisher Copyright: {\textcopyright} Proceedings of the 35th Pacific Asia Conference on Language, Information and Computation, PACLIC 2021.; 35th Pacific Asia Conference on Language, Information and Computation, PACLIC 2021 ; Conference date: 05-11-2021 Through 07-11-2021",

year = "2021",

language = "English",

pages = "436--445",

}

TY - CONF

T1 - Covering a sentence in form and meaning with fewer retrieved sentences

AU - Liu, Yuan

AU - Lepage, Yves

N1 - Funding Information: This research was supported in part by two grants-in-aid from the Japanese Society for the Promotion of Science (JSPS): no 18K11447 entitled “Self-explainable and fast-to-train example-based machine translation using neural networks” and no 18K11446 entitled “Natural language processing for academic writing in English” . Publisher Copyright: © Proceedings of the 35th Pacific Asia Conference on Language, Information and Computation, PACLIC 2021.

PY - 2021

Y1 - 2021

N2 - Retrieving similar sentences from a given collection of sentences is essential in a range of applications. In this work, we propose a novel method to retrieve several sentences that cover an input sentence in form and meaning with minimal redundancy, so as to enhance the overall coverage quality of the output sentences. We focus on the hierarchical granularity levels of sentence pieces, matching from common or similar n-grams to finer-grained words o subwords, using techniques from similar sentence retrieval and monolingual phrase alignment. Our method shows promising source and target coverage evaluation results when applied to parallel corpora. This shows the potential of our approach if integrated into an example-based machine translation system.

AB - Retrieving similar sentences from a given collection of sentences is essential in a range of applications. In this work, we propose a novel method to retrieve several sentences that cover an input sentence in form and meaning with minimal redundancy, so as to enhance the overall coverage quality of the output sentences. We focus on the hierarchical granularity levels of sentence pieces, matching from common or similar n-grams to finer-grained words o subwords, using techniques from similar sentence retrieval and monolingual phrase alignment. Our method shows promising source and target coverage evaluation results when applied to parallel corpora. This shows the potential of our approach if integrated into an example-based machine translation system.

UR - http://www.scopus.com/inward/record.url?scp=85127512763&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85127512763&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85127512763

SP - 436

EP - 445

T2 - 35th Pacific Asia Conference on Language, Information and Computation, PACLIC 2021

Y2 - 5 November 2021 through 7 November 2021

ER -

Covering a sentence in form and meaning with fewer retrieved sentences

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this