Using Multi-Label Classification for Improved Question Answering

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschung

Standard

Using Multi-Label Classification for Improved Question Answering. / Usbeck, Ricardo; Hoffmann, Michael; Röder, Michael et al.
Conference XXX. 2017.

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschung

Harvard

Usbeck, R, Hoffmann, M, Röder, M, Lehmann, J & Ngomo, A-CN 2017, Using Multi-Label Classification for Improved Question Answering. in Conference XXX. https://doi.org/10.48550/arXiv.1710.08634

APA

Usbeck, R., Hoffmann, M., Röder, M., Lehmann, J., & Ngomo, A.-C. N. (2017). Using Multi-Label Classification for Improved Question Answering. Manuskript in Vorbereitung. In Conference XXX https://doi.org/10.48550/arXiv.1710.08634

Vancouver

Usbeck R, Hoffmann M, Röder M, Lehmann J, Ngomo ACN. Using Multi-Label Classification for Improved Question Answering. in Conference XXX. 2017 doi: 10.48550/arXiv.1710.08634

Bibtex

@inbook{962878965c9749a39486ae8646875085,
title = "Using Multi-Label Classification for Improved Question Answering",
abstract = " A plethora of diverse approaches for question answering over RDF data have been developed in recent years. While the accuracy of these systems has increased significantly over time, most systems still focus on particular types of questions or particular challenges in question answering. What is a curse for single systems is a blessing for the combination of these systems. We show in this paper how machine learning techniques can be applied to create a more accurate question answering metasystem by reusing existing systems. In particular, we develop a multi-label classification-based metasystem for question answering over 6 existing systems using an innovative set of 14 question features. The metasystem outperforms the best single system by 14% F-measure on the recent QALD-6 benchmark. Furthermore, we analyzed the influence and correlation of the underlying features on the metasystem quality. ",
keywords = "cs.IR, cs.CL, Informatics",
author = "Ricardo Usbeck and Michael Hoffmann and Michael R{\"o}der and Jens Lehmann and Ngomo, {Axel-Cyrille Ngonga}",
note = "15 pages, 4 Tables, 3 Figues",
year = "2017",
month = oct,
day = "24",
doi = "10.48550/arXiv.1710.08634",
language = "English",
booktitle = "Conference XXX",

}

RIS

TY - CHAP

T1 - Using Multi-Label Classification for Improved Question Answering

AU - Usbeck, Ricardo

AU - Hoffmann, Michael

AU - Röder, Michael

AU - Lehmann, Jens

AU - Ngomo, Axel-Cyrille Ngonga

N1 - 15 pages, 4 Tables, 3 Figues

PY - 2017/10/24

Y1 - 2017/10/24

N2 - A plethora of diverse approaches for question answering over RDF data have been developed in recent years. While the accuracy of these systems has increased significantly over time, most systems still focus on particular types of questions or particular challenges in question answering. What is a curse for single systems is a blessing for the combination of these systems. We show in this paper how machine learning techniques can be applied to create a more accurate question answering metasystem by reusing existing systems. In particular, we develop a multi-label classification-based metasystem for question answering over 6 existing systems using an innovative set of 14 question features. The metasystem outperforms the best single system by 14% F-measure on the recent QALD-6 benchmark. Furthermore, we analyzed the influence and correlation of the underlying features on the metasystem quality.

AB - A plethora of diverse approaches for question answering over RDF data have been developed in recent years. While the accuracy of these systems has increased significantly over time, most systems still focus on particular types of questions or particular challenges in question answering. What is a curse for single systems is a blessing for the combination of these systems. We show in this paper how machine learning techniques can be applied to create a more accurate question answering metasystem by reusing existing systems. In particular, we develop a multi-label classification-based metasystem for question answering over 6 existing systems using an innovative set of 14 question features. The metasystem outperforms the best single system by 14% F-measure on the recent QALD-6 benchmark. Furthermore, we analyzed the influence and correlation of the underlying features on the metasystem quality.

KW - cs.IR

KW - cs.CL

KW - Informatics

U2 - 10.48550/arXiv.1710.08634

DO - 10.48550/arXiv.1710.08634

M3 - Article in conference proceedings

BT - Conference XXX

ER -

DOI