A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization

Yuliska; Tetsuya Sakai

doi:10.1109/INFOCT.2019.8710851

A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization

Yuliska, Tetsuya Sakai

School of Fundamental Science and Engineering

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

Abstract

Query-focused multi-document summarization aims to produce a single, short document that summarizes a set of documents that are relevant to a given query. During the past few years, deep learning approaches have been utilized to generate summaries in an abstractive or extractive manner. In this study, we employ six deep neural network approaches to solving a query-focused extractive multi-document summarization task and compare their performances. To the best of our knowledge, our study is the first to compare deep learning techniques on extractive query-focused multi-document summarization. Our experiments with DUC 2005-2007 benchmark datasets show that Bi-LSTM with Max-pooling achieves the highest performance among the methods compared.

Original language	English
Title of host publication	2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	153-157
Number of pages	5
ISBN (Electronic)	9781728133232
DOIs	https://doi.org/10.1109/INFOCT.2019.8710851
Publication status	Published - 2019 May 9
Event	2nd IEEE International Conference on Information and Computer Technologies, ICICT 2019 - Kahului, United States Duration: 2019 Mar 14 → 2019 Mar 17

Publication series

Name	2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019

Conference

Conference	2nd IEEE International Conference on Information and Computer Technologies, ICICT 2019
Country	United States
City	Kahului
Period	19/3/14 → 19/3/17

Keywords

comparative study
deep neural network
extractive summarization
query-focused summarization
text summarization

ASJC Scopus subject areas

Artificial Intelligence
Social Sciences (miscellaneous)
Computer Networks and Communications
Computer Science Applications
Information Systems
Information Systems and Management

Access to Document

10.1109/INFOCT.2019.8710851

Fingerprint Dive into the research topics of 'A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization'. Together they form a unique fingerprint.

Cite this

Yuliska, & Sakai, T. (2019). A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization. In 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019 (pp. 153-157). [8710851] (2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INFOCT.2019.8710851

A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization. / Yuliska, ; Sakai, Tetsuya.

2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 153-157 8710851 (2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019).

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

Yuliska, & Sakai, T 2019, A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization. in 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019., 8710851, 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019, Institute of Electrical and Electronics Engineers Inc., pp. 153-157, 2nd IEEE International Conference on Information and Computer Technologies, ICICT 2019, Kahului, United States, 19/3/14. https://doi.org/10.1109/INFOCT.2019.8710851

Yuliska , Sakai T. A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization. In 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 153-157. 8710851. (2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019). https://doi.org/10.1109/INFOCT.2019.8710851

Yuliska, ; Sakai, Tetsuya. / A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization. 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 153-157 (2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019).

@inproceedings{54121f3464604894a5356c79460f415d,

title = "A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization",

abstract = "Query-focused multi-document summarization aims to produce a single, short document that summarizes a set of documents that are relevant to a given query. During the past few years, deep learning approaches have been utilized to generate summaries in an abstractive or extractive manner. In this study, we employ six deep neural network approaches to solving a query-focused extractive multi-document summarization task and compare their performances. To the best of our knowledge, our study is the first to compare deep learning techniques on extractive query-focused multi-document summarization. Our experiments with DUC 2005-2007 benchmark datasets show that Bi-LSTM with Max-pooling achieves the highest performance among the methods compared.",

keywords = "comparative study, deep neural network, extractive summarization, query-focused summarization, text summarization",

author = "Yuliska and Tetsuya Sakai",

year = "2019",

month = may,

day = "9",

doi = "10.1109/INFOCT.2019.8710851",

language = "English",

series = "2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "153--157",

booktitle = "2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019",

note = "2nd IEEE International Conference on Information and Computer Technologies, ICICT 2019 ; Conference date: 14-03-2019 Through 17-03-2019",

}

TY - GEN

T1 - A Comparative Study of Deep Learning Approaches for Query-Focused Extractive Multi-Document Summarization

AU - Yuliska,

AU - Sakai, Tetsuya

PY - 2019/5/9

Y1 - 2019/5/9

N2 - Query-focused multi-document summarization aims to produce a single, short document that summarizes a set of documents that are relevant to a given query. During the past few years, deep learning approaches have been utilized to generate summaries in an abstractive or extractive manner. In this study, we employ six deep neural network approaches to solving a query-focused extractive multi-document summarization task and compare their performances. To the best of our knowledge, our study is the first to compare deep learning techniques on extractive query-focused multi-document summarization. Our experiments with DUC 2005-2007 benchmark datasets show that Bi-LSTM with Max-pooling achieves the highest performance among the methods compared.

AB - Query-focused multi-document summarization aims to produce a single, short document that summarizes a set of documents that are relevant to a given query. During the past few years, deep learning approaches have been utilized to generate summaries in an abstractive or extractive manner. In this study, we employ six deep neural network approaches to solving a query-focused extractive multi-document summarization task and compare their performances. To the best of our knowledge, our study is the first to compare deep learning techniques on extractive query-focused multi-document summarization. Our experiments with DUC 2005-2007 benchmark datasets show that Bi-LSTM with Max-pooling achieves the highest performance among the methods compared.

KW - comparative study

KW - deep neural network

KW - extractive summarization

KW - query-focused summarization

KW - text summarization

UR - http://www.scopus.com/inward/record.url?scp=85066608855&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066608855&partnerID=8YFLogxK

U2 - 10.1109/INFOCT.2019.8710851

DO - 10.1109/INFOCT.2019.8710851

M3 - Conference contribution

AN - SCOPUS:85066608855

T3 - 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019

SP - 153

EP - 157

BT - 2019 IEEE 2nd International Conference on Information and Computer Technologies, ICICT 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd IEEE International Conference on Information and Computer Technologies, ICICT 2019

Y2 - 14 March 2019 through 17 March 2019

ER -