Exploring differential topic models for comparative summarization of scientific papers

Lei He; Wei Li; Hai Zhuge

Exploring differential topic models for comparative summarization of scientific papers

Lei He, Wei Li, Hai Zhuge

Computer Science Research Group

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

Abstract

This paper investigates differential topic models (dTM) for summarizing the differences among document groups. Starting from a simple probabilistic generative model, we propose dTM-SAGE that explicitly models the deviations on group-specific word distributions to indicate how words are used differentially across different document groups from a background word distribution. It is more effective to capture unique characteristics for comparing document groups. To generate dTM-based comparative summaries, we propose two sentence scoring methods for measuring the sentence discriminative capacity. Experimental results on scientific papers dataset show that our dTM-based comparative summarization methods significantly outperform the generic baselines and the state-of-the-art comparative summarization
methods under ROUGE metrics.

Original language	English
Title of host publication	Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics
Subtitle of host publication	technical papers
Publisher	Association for Computational Linguistics
Pages	1028-1038
Number of pages	10
ISBN (Print)	978-4-87974-702-0
Publication status	Published - 11 Dec 2016
Event	26th International Conference on Computational Linguistics: COLIN 2016 - Osaka, Japan Duration: 11 Dec 2016 → 16 Dec 2016 Conference number: 26

Conference

Conference	26th International Conference on Computational Linguistics
Abbreviated title	COLING 2016
Country/Territory	Japan
City	Osaka
Period	11/12/16 → 16/12/16

Bibliographical note

-This work is licenced under a Creative Commons Attribution 4.0 International License. License details: http:// creativecommons.org/licenses/by/4.0/

Access to Document

Exploring differential topic models for comparative summarization of scientific papers
This work is licenced under a Creative Commons Attribution 4.0 International License. License details: http://creativecommons. org/licenses/by/4.0/
Final published version, 459 KBLicence: CC BY 3.0

Cite this

@inproceedings{3ae749bda88e49e7aedd7fe963e0b842,

title = "Exploring differential topic models for comparative summarization of scientific papers",

abstract = "This paper investigates differential topic models (dTM) for summarizing the differences among document groups. Starting from a simple probabilistic generative model, we propose dTM-SAGE that explicitly models the deviations on group-specific word distributions to indicate how words are used differentially across different document groups from a background word distribution. It is more effective to capture unique characteristics for comparing document groups. To generate dTM-based comparative summaries, we propose two sentence scoring methods for measuring the sentence discriminative capacity. Experimental results on scientific papers dataset show that our dTM-based comparative summarization methods significantly outperform the generic baselines and the state-of-the-art comparative summarizationmethods under ROUGE metrics.",

author = "Lei He and Wei Li and Hai Zhuge",

note = "-This work is licenced under a Creative Commons Attribution 4.0 International License. License details: http:// creativecommons.org/licenses/by/4.0/; 26th International Conference on Computational Linguistics : COLIN 2016, COLING 2016 ; Conference date: 11-12-2016 Through 16-12-2016",

year = "2016",

month = dec,

day = "11",

language = "English",

isbn = "978-4-87974-702-0",

pages = "1028--1038",

booktitle = "Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics",

publisher = "Association for Computational Linguistics",

}

He, L, Li, W & Zhuge, H 2016, Exploring differential topic models for comparative summarization of scientific papers. in Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: technical papers., C16-1098, Association for Computational Linguistics, pp. 1028-1038, 26th International Conference on Computational Linguistics, Osaka, Japan, 11/12/16.

Exploring differential topic models for comparative summarization of scientific papers. / He, Lei; Li, Wei; Zhuge, Hai.
Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: technical papers. Association for Computational Linguistics, 2016. p. 1028-1038 C16-1098.

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

TY - GEN

T1 - Exploring differential topic models for comparative summarization of scientific papers

AU - He, Lei

AU - Li, Wei

AU - Zhuge, Hai

N1 - Conference code: 26

PY - 2016/12/11

Y1 - 2016/12/11

N2 - This paper investigates differential topic models (dTM) for summarizing the differences among document groups. Starting from a simple probabilistic generative model, we propose dTM-SAGE that explicitly models the deviations on group-specific word distributions to indicate how words are used differentially across different document groups from a background word distribution. It is more effective to capture unique characteristics for comparing document groups. To generate dTM-based comparative summaries, we propose two sentence scoring methods for measuring the sentence discriminative capacity. Experimental results on scientific papers dataset show that our dTM-based comparative summarization methods significantly outperform the generic baselines and the state-of-the-art comparative summarizationmethods under ROUGE metrics.

AB - This paper investigates differential topic models (dTM) for summarizing the differences among document groups. Starting from a simple probabilistic generative model, we propose dTM-SAGE that explicitly models the deviations on group-specific word distributions to indicate how words are used differentially across different document groups from a background word distribution. It is more effective to capture unique characteristics for comparing document groups. To generate dTM-based comparative summaries, we propose two sentence scoring methods for measuring the sentence discriminative capacity. Experimental results on scientific papers dataset show that our dTM-based comparative summarization methods significantly outperform the generic baselines and the state-of-the-art comparative summarizationmethods under ROUGE metrics.

UR - http://proceedings.aom.org/content/2016/1/15978

M3 - Conference publication

SN - 978-4-87974-702-0

SP - 1028

EP - 1038

BT - Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics

PB - Association for Computational Linguistics

T2 - 26th International Conference on Computational Linguistics

Y2 - 11 December 2016 through 16 December 2016

ER -

Exploring differential topic models for comparative summarization of scientific papers

Abstract

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this