A syntax-aware re-ranker for microblog retrieval

Aliaksei Severyn, Alessandro Moschitti, Manos Tsagkias, Richard Berendsen, Maarten De Rijke

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

14 Citations (Scopus)

Abstract

We tackle the problem of improving microblog retrieval algorithms by proposing a robust structural representation of (query, tweet) pairs. We employ these structures in a principled kernel learning framework that automatically extracts and learns highly discriminative features. We test the generalization power of our approach on the TREC Microblog 2011 and 2012 tasks. We find that relational syntactic features generated by structural kernels are effective for learning to rank (L2R) and can easily be combined with those of other existing systems to boost their accuracy. In particular, the results show that our L2R approach improves on almost all the participating systems at TREC, only using their raw scores as a single feature. Our method yields an average increase of 5% in retrieval effectiveness and 7 positions in system ranks.

Original languageEnglish
Title of host publicationSIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval
PublisherAssociation for Computing Machinery
Pages1067-1070
Number of pages4
ISBN (Print)9781450322591
DOIs
Publication statusPublished - 1 Jan 2014
Event37th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2014 - Gold Coast, QLD, Australia
Duration: 6 Jul 201411 Jul 2014

Other

Other37th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2014
CountryAustralia
CityGold Coast, QLD
Period6/7/1411/7/14

Fingerprint

Syntactics

Keywords

  • Microblog search
  • Re-ranking
  • Semantic modeling

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Information Systems

Cite this

Severyn, A., Moschitti, A., Tsagkias, M., Berendsen, R., & De Rijke, M. (2014). A syntax-aware re-ranker for microblog retrieval. In SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 1067-1070). Association for Computing Machinery. https://doi.org/10.1145/2600428.2609511

A syntax-aware re-ranker for microblog retrieval. / Severyn, Aliaksei; Moschitti, Alessandro; Tsagkias, Manos; Berendsen, Richard; De Rijke, Maarten.

SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, 2014. p. 1067-1070.

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

Severyn, A, Moschitti, A, Tsagkias, M, Berendsen, R & De Rijke, M 2014, A syntax-aware re-ranker for microblog retrieval. in SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, pp. 1067-1070, 37th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2014, Gold Coast, QLD, Australia, 6/7/14. https://doi.org/10.1145/2600428.2609511
Severyn A, Moschitti A, Tsagkias M, Berendsen R, De Rijke M. A syntax-aware re-ranker for microblog retrieval. In SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery. 2014. p. 1067-1070 https://doi.org/10.1145/2600428.2609511
Severyn, Aliaksei ; Moschitti, Alessandro ; Tsagkias, Manos ; Berendsen, Richard ; De Rijke, Maarten. / A syntax-aware re-ranker for microblog retrieval. SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, 2014. pp. 1067-1070
@inproceedings{16b9f4795bb04fc9951d5f9911124c5b,
title = "A syntax-aware re-ranker for microblog retrieval",
abstract = "We tackle the problem of improving microblog retrieval algorithms by proposing a robust structural representation of (query, tweet) pairs. We employ these structures in a principled kernel learning framework that automatically extracts and learns highly discriminative features. We test the generalization power of our approach on the TREC Microblog 2011 and 2012 tasks. We find that relational syntactic features generated by structural kernels are effective for learning to rank (L2R) and can easily be combined with those of other existing systems to boost their accuracy. In particular, the results show that our L2R approach improves on almost all the participating systems at TREC, only using their raw scores as a single feature. Our method yields an average increase of 5{\%} in retrieval effectiveness and 7 positions in system ranks.",
keywords = "Microblog search, Re-ranking, Semantic modeling",
author = "Aliaksei Severyn and Alessandro Moschitti and Manos Tsagkias and Richard Berendsen and {De Rijke}, Maarten",
year = "2014",
month = "1",
day = "1",
doi = "10.1145/2600428.2609511",
language = "English",
isbn = "9781450322591",
pages = "1067--1070",
booktitle = "SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - A syntax-aware re-ranker for microblog retrieval

AU - Severyn, Aliaksei

AU - Moschitti, Alessandro

AU - Tsagkias, Manos

AU - Berendsen, Richard

AU - De Rijke, Maarten

PY - 2014/1/1

Y1 - 2014/1/1

N2 - We tackle the problem of improving microblog retrieval algorithms by proposing a robust structural representation of (query, tweet) pairs. We employ these structures in a principled kernel learning framework that automatically extracts and learns highly discriminative features. We test the generalization power of our approach on the TREC Microblog 2011 and 2012 tasks. We find that relational syntactic features generated by structural kernels are effective for learning to rank (L2R) and can easily be combined with those of other existing systems to boost their accuracy. In particular, the results show that our L2R approach improves on almost all the participating systems at TREC, only using their raw scores as a single feature. Our method yields an average increase of 5% in retrieval effectiveness and 7 positions in system ranks.

AB - We tackle the problem of improving microblog retrieval algorithms by proposing a robust structural representation of (query, tweet) pairs. We employ these structures in a principled kernel learning framework that automatically extracts and learns highly discriminative features. We test the generalization power of our approach on the TREC Microblog 2011 and 2012 tasks. We find that relational syntactic features generated by structural kernels are effective for learning to rank (L2R) and can easily be combined with those of other existing systems to boost their accuracy. In particular, the results show that our L2R approach improves on almost all the participating systems at TREC, only using their raw scores as a single feature. Our method yields an average increase of 5% in retrieval effectiveness and 7 positions in system ranks.

KW - Microblog search

KW - Re-ranking

KW - Semantic modeling

UR - http://www.scopus.com/inward/record.url?scp=84904544588&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904544588&partnerID=8YFLogxK

U2 - 10.1145/2600428.2609511

DO - 10.1145/2600428.2609511

M3 - Conference contribution

SN - 9781450322591

SP - 1067

EP - 1070

BT - SIGIR 2014 - Proceedings of the 37th International ACM SIGIR Conference on Research and Development in Information Retrieval

PB - Association for Computing Machinery

ER -