Why Reinvent the Wheel: Let's Build Question Answering Systems Together

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

  • Kuldeep Singh
  • Arun Sethupat Radhakrishna
  • Andreas Both
  • Saeedeh Shekarpour
  • Ioanna Lytra
  • Ricardo Usbeck
  • Akhilesh Vyas
  • Akmal Khikmatullaev
  • Dharmen Punjani
  • Christoph Lange
  • Maria Esther Vidal
  • Jens Lehmann
  • Soeren Auer

External Research Organisations

  • University of Bonn
  • Fraunhofer Institute for Intelligent Analysis and Information Systems (IAIS)
  • University of Minnesota
  • DATEV eG
  • University of Dayton
  • Paderborn University
  • University of Athens
  • German National Library of Science and Technology (TIB)
View graph of relations

Details

Original languageEnglish
Title of host publicationThe Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018
Subtitle of host publicationProceedings of the 2018 World Wide Web Conference
Place of PublicationGeneva
PublisherAssociation for Computing Machinery (ACM)
Pages1247-1256
Number of pages10
ISBN (electronic)9781450356398
Publication statusPublished - 10 Apr 2018

Publication series

NameProceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18

Abstract

Modern question answering (QA) systems need to flexibly integrate a number of components specialised to fulfil specific tasks in a QA pipeline. Key QA tasks include Named Entity Recognition and Disambiguation, Relation Extraction, and Query Building. Since a number of different software components exist that implement different strategies for each of these tasks, it is a major challenge to select and combine the most suitable components into a QA system, given the characteristics of a question. We study this optimisation problem and train classifiers, which take features of a question as input and have the goal of optimising the selection of QA components based on those features. We then devise a greedy algorithm to identify the pipelines that include the suitable components and can effectively answer the given question. We implement this model within Frankenstein, a QA framework able to select QA components and compose QA pipelines. We evaluate the effectiveness of the pipelines generated by Frankenstein using the QALD and LC-QuAD benchmarks. These results not only suggest that Frankenstein precisely solves the QA optimisation problem but also enables the automatic composition of optimised QA pipelines, which outperform the static Baseline QA pipeline. Thanks to this flexible and fully automated pipeline generation process, new QA components can be easily included in Frankenstein, thus improving the performance of the generated pipelines.

Keywords

    Question Answering, Software Reusability, Semantic Web, Semantic Search, QA Framework, Semantic web, Semantic search, Question answering, QA framework, Software reusability

ASJC Scopus subject areas

Cite this

Why Reinvent the Wheel: Let's Build Question Answering Systems Together. / Singh, Kuldeep; Radhakrishna, Arun Sethupat; Both, Andreas et al.
The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018: Proceedings of the 2018 World Wide Web Conference. Geneva: Association for Computing Machinery (ACM), 2018. p. 1247-1256 (Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Singh, K, Radhakrishna, AS, Both, A, Shekarpour, S, Lytra, I, Usbeck, R, Vyas, A, Khikmatullaev, A, Punjani, D, Lange, C, Vidal, ME, Lehmann, J & Auer, S 2018, Why Reinvent the Wheel: Let's Build Question Answering Systems Together. in The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018: Proceedings of the 2018 World Wide Web Conference. Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18, Association for Computing Machinery (ACM), Geneva, pp. 1247-1256. https://doi.org/10.1145/3178876.3186023
Singh, K., Radhakrishna, A. S., Both, A., Shekarpour, S., Lytra, I., Usbeck, R., Vyas, A., Khikmatullaev, A., Punjani, D., Lange, C., Vidal, M. E., Lehmann, J., & Auer, S. (2018). Why Reinvent the Wheel: Let's Build Question Answering Systems Together. In The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018: Proceedings of the 2018 World Wide Web Conference (pp. 1247-1256). (Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18). Association for Computing Machinery (ACM). https://doi.org/10.1145/3178876.3186023
Singh K, Radhakrishna AS, Both A, Shekarpour S, Lytra I, Usbeck R et al. Why Reinvent the Wheel: Let's Build Question Answering Systems Together. In The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018: Proceedings of the 2018 World Wide Web Conference. Geneva: Association for Computing Machinery (ACM). 2018. p. 1247-1256. (Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18). doi: 10.1145/3178876.3186023
Singh, Kuldeep ; Radhakrishna, Arun Sethupat ; Both, Andreas et al. / Why Reinvent the Wheel : Let's Build Question Answering Systems Together. The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018: Proceedings of the 2018 World Wide Web Conference. Geneva : Association for Computing Machinery (ACM), 2018. pp. 1247-1256 (Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18).
Download
@inproceedings{f73e43feff014d7eaafa48cf9b2ec6a9,
title = "Why Reinvent the Wheel: Let's Build Question Answering Systems Together",
abstract = "Modern question answering (QA) systems need to flexibly integrate a number of components specialised to fulfil specific tasks in a QA pipeline. Key QA tasks include Named Entity Recognition and Disambiguation, Relation Extraction, and Query Building. Since a number of different software components exist that implement different strategies for each of these tasks, it is a major challenge to select and combine the most suitable components into a QA system, given the characteristics of a question. We study this optimisation problem and train classifiers, which take features of a question as input and have the goal of optimising the selection of QA components based on those features. We then devise a greedy algorithm to identify the pipelines that include the suitable components and can effectively answer the given question. We implement this model within Frankenstein, a QA framework able to select QA components and compose QA pipelines. We evaluate the effectiveness of the pipelines generated by Frankenstein using the QALD and LC-QuAD benchmarks. These results not only suggest that Frankenstein precisely solves the QA optimisation problem but also enables the automatic composition of optimised QA pipelines, which outperform the static Baseline QA pipeline. Thanks to this flexible and fully automated pipeline generation process, new QA components can be easily included in Frankenstein, thus improving the performance of the generated pipelines.",
keywords = "Question Answering, Software Reusability, Semantic Web, Semantic Search, QA Framework, Semantic web, Semantic search, Question answering, QA framework, Software reusability",
author = "Kuldeep Singh and Radhakrishna, {Arun Sethupat} and Andreas Both and Saeedeh Shekarpour and Ioanna Lytra and Ricardo Usbeck and Akhilesh Vyas and Akmal Khikmatullaev and Dharmen Punjani and Christoph Lange and Vidal, {Maria Esther} and Jens Lehmann and Soeren Auer",
note = "Funding information: This work has received funding from the EU H2020 R&I programme for the Marie Sk?odowska-Curie action WDAqua (GA No 642795), Eurostars project QAMEL (E!9725), and EU H2020 R&I HOBBIT (GA 688227). We thank Yakun Li, Osmar Zaiane, and Anant Gupta for their useful suggestions. This work has received funding from the EU H2020 R&I programme for the Marie Sklodowska-Curie action WDAqua (GA No 642795), Eurostars project QAMEL (E!9725), and EU H2020 R&I HOBBIT (GA 688227). We thank Yakun Li, Osmar Zaiane, and Anant Gupta for their useful suggestions.",
year = "2018",
month = apr,
day = "10",
doi = "10.1145/3178876.3186023",
language = "English",
series = "Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18",
publisher = "Association for Computing Machinery (ACM)",
pages = "1247--1256",
booktitle = "The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018",
address = "United States",

}

Download

TY - GEN

T1 - Why Reinvent the Wheel

T2 - Let's Build Question Answering Systems Together

AU - Singh, Kuldeep

AU - Radhakrishna, Arun Sethupat

AU - Both, Andreas

AU - Shekarpour, Saeedeh

AU - Lytra, Ioanna

AU - Usbeck, Ricardo

AU - Vyas, Akhilesh

AU - Khikmatullaev, Akmal

AU - Punjani, Dharmen

AU - Lange, Christoph

AU - Vidal, Maria Esther

AU - Lehmann, Jens

AU - Auer, Soeren

N1 - Funding information: This work has received funding from the EU H2020 R&I programme for the Marie Sk?odowska-Curie action WDAqua (GA No 642795), Eurostars project QAMEL (E!9725), and EU H2020 R&I HOBBIT (GA 688227). We thank Yakun Li, Osmar Zaiane, and Anant Gupta for their useful suggestions. This work has received funding from the EU H2020 R&I programme for the Marie Sklodowska-Curie action WDAqua (GA No 642795), Eurostars project QAMEL (E!9725), and EU H2020 R&I HOBBIT (GA 688227). We thank Yakun Li, Osmar Zaiane, and Anant Gupta for their useful suggestions.

PY - 2018/4/10

Y1 - 2018/4/10

N2 - Modern question answering (QA) systems need to flexibly integrate a number of components specialised to fulfil specific tasks in a QA pipeline. Key QA tasks include Named Entity Recognition and Disambiguation, Relation Extraction, and Query Building. Since a number of different software components exist that implement different strategies for each of these tasks, it is a major challenge to select and combine the most suitable components into a QA system, given the characteristics of a question. We study this optimisation problem and train classifiers, which take features of a question as input and have the goal of optimising the selection of QA components based on those features. We then devise a greedy algorithm to identify the pipelines that include the suitable components and can effectively answer the given question. We implement this model within Frankenstein, a QA framework able to select QA components and compose QA pipelines. We evaluate the effectiveness of the pipelines generated by Frankenstein using the QALD and LC-QuAD benchmarks. These results not only suggest that Frankenstein precisely solves the QA optimisation problem but also enables the automatic composition of optimised QA pipelines, which outperform the static Baseline QA pipeline. Thanks to this flexible and fully automated pipeline generation process, new QA components can be easily included in Frankenstein, thus improving the performance of the generated pipelines.

AB - Modern question answering (QA) systems need to flexibly integrate a number of components specialised to fulfil specific tasks in a QA pipeline. Key QA tasks include Named Entity Recognition and Disambiguation, Relation Extraction, and Query Building. Since a number of different software components exist that implement different strategies for each of these tasks, it is a major challenge to select and combine the most suitable components into a QA system, given the characteristics of a question. We study this optimisation problem and train classifiers, which take features of a question as input and have the goal of optimising the selection of QA components based on those features. We then devise a greedy algorithm to identify the pipelines that include the suitable components and can effectively answer the given question. We implement this model within Frankenstein, a QA framework able to select QA components and compose QA pipelines. We evaluate the effectiveness of the pipelines generated by Frankenstein using the QALD and LC-QuAD benchmarks. These results not only suggest that Frankenstein precisely solves the QA optimisation problem but also enables the automatic composition of optimised QA pipelines, which outperform the static Baseline QA pipeline. Thanks to this flexible and fully automated pipeline generation process, new QA components can be easily included in Frankenstein, thus improving the performance of the generated pipelines.

KW - Question Answering

KW - Software Reusability

KW - Semantic Web

KW - Semantic Search

KW - QA Framework

KW - Semantic web

KW - Semantic search

KW - Question answering

KW - QA framework

KW - Software reusability

UR - http://www.scopus.com/inward/record.url?scp=85076220402&partnerID=8YFLogxK

U2 - 10.1145/3178876.3186023

DO - 10.1145/3178876.3186023

M3 - Conference contribution

T3 - Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18

SP - 1247

EP - 1256

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

PB - Association for Computing Machinery (ACM)

CY - Geneva

ER -

By the same author(s)