A Neural Network-based Framework for Non-factoid Question Answering

Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review

Autoren

  • Nam Khanh Tran
  • Claudia Niederée

Organisationseinheiten

Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Titel des SammelwerksThe Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018
Herausgeber (Verlag)Association for Computing Machinery (ACM)
Seiten1979-1983
Seitenumfang5
ISBN (elektronisch)9781450356404
PublikationsstatusVeröffentlicht - 23 Apr. 2018
Veranstaltung27th International World Wide Web, WWW 2018 - Lyon, Frankreich
Dauer: 23 Apr. 2018 → 27 Apr. 2018

Publikationsreihe

NameThe Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018
Band2018-January

Abstract

In this paper, we present a neural network based framework for answering non-factoid questions. The framework consists of two main components: Answer Retriever and Answer Ranker. In the first component, we leverage off-the-shelf retrieval models (e.g. bm25) to retrieve a pool of candidate answers regarding to the input question. Answer Ranker is then used to select the most suitable answer. In this work, we adopt two typical deep learning based frameworks for our Answer Ranker component. One is based on Siamese architecture and the other is the Compare-Aggregate framework. The Answer Ranker component is evaluated separately based on popular answer selection datasets. Our overall system is evaluated using FiQA dataset, a newly released dataset for financial domain and shows promising results.

ASJC Scopus Sachgebiete

Zitieren

A Neural Network-based Framework for Non-factoid Question Answering. / Tran, Nam Khanh; Niederée, Claudia.
The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018. Association for Computing Machinery (ACM), 2018. S. 1979-1983 (The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018; Band 2018-January).

Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review

Tran, NK & Niederée, C 2018, A Neural Network-based Framework for Non-factoid Question Answering. in The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018. The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018, Bd. 2018-January, Association for Computing Machinery (ACM), S. 1979-1983, 27th International World Wide Web, WWW 2018, Lyon, Frankreich, 23 Apr. 2018. https://doi.org/10.1145/3184558.3191830
Tran, N. K., & Niederée, C. (2018). A Neural Network-based Framework for Non-factoid Question Answering. In The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018 (S. 1979-1983). (The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018; Band 2018-January). Association for Computing Machinery (ACM). https://doi.org/10.1145/3184558.3191830
Tran NK, Niederée C. A Neural Network-based Framework for Non-factoid Question Answering. in The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018. Association for Computing Machinery (ACM). 2018. S. 1979-1983. (The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018). doi: 10.1145/3184558.3191830
Tran, Nam Khanh ; Niederée, Claudia. / A Neural Network-based Framework for Non-factoid Question Answering. The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018. Association for Computing Machinery (ACM), 2018. S. 1979-1983 (The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018).
Download
@inproceedings{7fb0abfcafd44391bf6f5ebbf2fdefc8,
title = "A Neural Network-based Framework for Non-factoid Question Answering",
abstract = "In this paper, we present a neural network based framework for answering non-factoid questions. The framework consists of two main components: Answer Retriever and Answer Ranker. In the first component, we leverage off-the-shelf retrieval models (e.g. bm25) to retrieve a pool of candidate answers regarding to the input question. Answer Ranker is then used to select the most suitable answer. In this work, we adopt two typical deep learning based frameworks for our Answer Ranker component. One is based on Siamese architecture and the other is the Compare-Aggregate framework. The Answer Ranker component is evaluated separately based on popular answer selection datasets. Our overall system is evaluated using FiQA dataset, a newly released dataset for financial domain and shows promising results.",
keywords = "Non-factoid question answering, representation learning",
author = "Tran, {Nam Khanh} and Claudia Nieder{\'e}e",
note = "Funding Information: This work was partially funded by the German Federal Ministry of Education and Research (BMBF) for the project eLabour (01UG1512C). ; 27th International World Wide Web, WWW 2018 ; Conference date: 23-04-2018 Through 27-04-2018",
year = "2018",
month = apr,
day = "23",
doi = "10.1145/3184558.3191830",
language = "English",
series = "The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018",
publisher = "Association for Computing Machinery (ACM)",
pages = "1979--1983",
booktitle = "The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018",
address = "United States",

}

Download

TY - GEN

T1 - A Neural Network-based Framework for Non-factoid Question Answering

AU - Tran, Nam Khanh

AU - Niederée, Claudia

N1 - Funding Information: This work was partially funded by the German Federal Ministry of Education and Research (BMBF) for the project eLabour (01UG1512C).

PY - 2018/4/23

Y1 - 2018/4/23

N2 - In this paper, we present a neural network based framework for answering non-factoid questions. The framework consists of two main components: Answer Retriever and Answer Ranker. In the first component, we leverage off-the-shelf retrieval models (e.g. bm25) to retrieve a pool of candidate answers regarding to the input question. Answer Ranker is then used to select the most suitable answer. In this work, we adopt two typical deep learning based frameworks for our Answer Ranker component. One is based on Siamese architecture and the other is the Compare-Aggregate framework. The Answer Ranker component is evaluated separately based on popular answer selection datasets. Our overall system is evaluated using FiQA dataset, a newly released dataset for financial domain and shows promising results.

AB - In this paper, we present a neural network based framework for answering non-factoid questions. The framework consists of two main components: Answer Retriever and Answer Ranker. In the first component, we leverage off-the-shelf retrieval models (e.g. bm25) to retrieve a pool of candidate answers regarding to the input question. Answer Ranker is then used to select the most suitable answer. In this work, we adopt two typical deep learning based frameworks for our Answer Ranker component. One is based on Siamese architecture and the other is the Compare-Aggregate framework. The Answer Ranker component is evaluated separately based on popular answer selection datasets. Our overall system is evaluated using FiQA dataset, a newly released dataset for financial domain and shows promising results.

KW - Non-factoid question answering

KW - representation learning

UR - http://www.scopus.com/inward/record.url?scp=85110423076&partnerID=8YFLogxK

U2 - 10.1145/3184558.3191830

DO - 10.1145/3184558.3191830

M3 - Conference contribution

AN - SCOPUS:85110423076

T3 - The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018

SP - 1979

EP - 1983

BT - The Web Conference 2018 - Companion of the World Wide Web Conference, WWW 2018

PB - Association for Computing Machinery (ACM)

T2 - 27th International World Wide Web, WWW 2018

Y2 - 23 April 2018 through 27 April 2018

ER -