Controlling information aggregation for complex question answering

Heeyoung Kwon, Harsh Trivedi, Peter Jansen, Mihai Surdeanu, Niranjan Balasubramanian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Complex question answering, the task of answering complex natural language questions that rely on inference, requires the aggregation of information from multiple sources. Automatic aggregation often fails because it combines semantically unrelated facts leading to bad inferences. This paper proposes methods to address this inference drift problem. In particular, the paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Empirical evaluation on an elementary science exam benchmark shows that the proposed methods enables effective aggregation even over larger graphs and demonstrates the complementary value of information aggregation for answering complex questions.

Original languageEnglish (US)
Title of host publicationAdvances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings
PublisherSpringer-Verlag
Pages750-757
Number of pages8
ISBN (Print)9783319769400
DOIs
StatePublished - Jan 1 2018
Event40th European Conference on Information Retrieval, ECIR 2018 - Grenoble, France
Duration: Mar 26 2018Mar 29 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10772 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other40th European Conference on Information Retrieval, ECIR 2018
CountryFrance
CityGrenoble
Period3/26/183/29/18

Fingerprint

Question Answering
Aggregation
Agglomeration
Value of Information
Information Extraction
Graph in graph theory
Natural Language
Random walk
Benchmark
Evaluation
Demonstrate

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Kwon, H., Trivedi, H., Jansen, P., Surdeanu, M., & Balasubramanian, N. (2018). Controlling information aggregation for complex question answering. In Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings (pp. 750-757). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10772 LNCS). Springer-Verlag. https://doi.org/10.1007/978-3-319-76941-7_72

Controlling information aggregation for complex question answering. / Kwon, Heeyoung; Trivedi, Harsh; Jansen, Peter; Surdeanu, Mihai; Balasubramanian, Niranjan.

Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings. Springer-Verlag, 2018. p. 750-757 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10772 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kwon, H, Trivedi, H, Jansen, P, Surdeanu, M & Balasubramanian, N 2018, Controlling information aggregation for complex question answering. in Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10772 LNCS, Springer-Verlag, pp. 750-757, 40th European Conference on Information Retrieval, ECIR 2018, Grenoble, France, 3/26/18. https://doi.org/10.1007/978-3-319-76941-7_72
Kwon H, Trivedi H, Jansen P, Surdeanu M, Balasubramanian N. Controlling information aggregation for complex question answering. In Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings. Springer-Verlag. 2018. p. 750-757. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-76941-7_72
Kwon, Heeyoung ; Trivedi, Harsh ; Jansen, Peter ; Surdeanu, Mihai ; Balasubramanian, Niranjan. / Controlling information aggregation for complex question answering. Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings. Springer-Verlag, 2018. pp. 750-757 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b69322bdf7324426a80f7c826fffcf13,
title = "Controlling information aggregation for complex question answering",
abstract = "Complex question answering, the task of answering complex natural language questions that rely on inference, requires the aggregation of information from multiple sources. Automatic aggregation often fails because it combines semantically unrelated facts leading to bad inferences. This paper proposes methods to address this inference drift problem. In particular, the paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Empirical evaluation on an elementary science exam benchmark shows that the proposed methods enables effective aggregation even over larger graphs and demonstrates the complementary value of information aggregation for answering complex questions.",
author = "Heeyoung Kwon and Harsh Trivedi and Peter Jansen and Mihai Surdeanu and Niranjan Balasubramanian",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-319-76941-7_72",
language = "English (US)",
isbn = "9783319769400",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "750--757",
booktitle = "Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings",

}

TY - GEN

T1 - Controlling information aggregation for complex question answering

AU - Kwon, Heeyoung

AU - Trivedi, Harsh

AU - Jansen, Peter

AU - Surdeanu, Mihai

AU - Balasubramanian, Niranjan

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Complex question answering, the task of answering complex natural language questions that rely on inference, requires the aggregation of information from multiple sources. Automatic aggregation often fails because it combines semantically unrelated facts leading to bad inferences. This paper proposes methods to address this inference drift problem. In particular, the paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Empirical evaluation on an elementary science exam benchmark shows that the proposed methods enables effective aggregation even over larger graphs and demonstrates the complementary value of information aggregation for answering complex questions.

AB - Complex question answering, the task of answering complex natural language questions that rely on inference, requires the aggregation of information from multiple sources. Automatic aggregation often fails because it combines semantically unrelated facts leading to bad inferences. This paper proposes methods to address this inference drift problem. In particular, the paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Empirical evaluation on an elementary science exam benchmark shows that the proposed methods enables effective aggregation even over larger graphs and demonstrates the complementary value of information aggregation for answering complex questions.

UR - http://www.scopus.com/inward/record.url?scp=85044477982&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044477982&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-76941-7_72

DO - 10.1007/978-3-319-76941-7_72

M3 - Conference contribution

AN - SCOPUS:85044477982

SN - 9783319769400

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 750

EP - 757

BT - Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings

PB - Springer-Verlag

ER -