Why reinvent the wheel: Let's build question answering systems together
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Standard
The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018. ed. / Pierre-Antoine Champin; Fabien Gandon; Lionel Medini. Association for Computing Machinery, Inc, 2018. p. 1247-1256 (The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018).
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Why reinvent the wheel
T2 - 27th International World Wide Web, WWW 2018
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, Sören
N1 - This work has received funding from the EU H2020 R&I programme for the Marie Skłodowska-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. Publisher Copyright: © 2018 IW3C2 (International World Wide Web Conference Committee), published under Creative Commons CC BY 4.0 License.
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 - QA framework
KW - Question answering
KW - Semantic search
KW - Semantic web
KW - Software reusability
KW - Informatics
KW - Business informatics
UR - http://www.scopus.com/inward/record.url?scp=85076220402&partnerID=8YFLogxK
U2 - 10.1145/3178876.3186023
DO - 10.1145/3178876.3186023
M3 - Article in conference proceedings
AN - SCOPUS:85076220402
SN - 978-1-4503-5639-8
T3 - The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018
SP - 1247
EP - 1256
BT - The Web Conference 2018 - Proceedings of the World Wide Web Conference, WWW 2018
A2 - Champin, Pierre-Antoine
A2 - Gandon, Fabien
A2 - Medini, Lionel
PB - Association for Computing Machinery, Inc
Y2 - 23 April 2018 through 27 April 2018
ER -