Ranking model selection and fusion for effective microblog search

Zhongyu Wei, Wei Gao, Tarek El-Ganainy, Walid Magdy, Kam Fai Wong

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

8 Citations (Scopus)

Abstract

Re-ranking was shown to have positive impact on the effectiveness for microblog search. Yet existing approaches mostly focused on using a single ranker to learn some better ranking function with respect to various relevance features. Given various available rank learners (such as learning to rank algorithms), in this work, we mainly study an orthogonal problem where multiple learned ranking models form an ensemble for re-ranking the retrieved tweets than just using a single ranking model in order to achieve higher search effectiveness. We explore the use of query-sensitive model selection and rank fusion methods based on the result lists produced from multiple rank learners. Base on the TREC microblog datasets, we found that our selection-based ensemble approach can significantly outperform using the single best ranker, and it also has clear advantage over the rank fusion that combines the results of all the available models.

Original languageEnglish
Title of host publicationSoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014
PublisherAssociation for Computing Machinery, Inc
Pages21-26
Number of pages6
ISBN (Print)9781450330220
DOIs
Publication statusPublished - 1 Jan 2014
Event1st ACM International Workshop on Social Media Retrieval and Analysis, SoMeRA 2014 - Co-located with SIGIR 2014 - Gold Coast, Australia
Duration: 11 Jul 201411 Jul 2014

Other

Other1st ACM International Workshop on Social Media Retrieval and Analysis, SoMeRA 2014 - Co-located with SIGIR 2014
CountryAustralia
CityGold Coast
Period11/7/1411/7/14

Fingerprint

ranking
Fusion reactions
learning

Keywords

  • Aggregation
  • Microblog search
  • Rank fusion
  • Ranker selection
  • Re-ranking
  • Twitter

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Media Technology
  • Communication

Cite this

Wei, Z., Gao, W., El-Ganainy, T., Magdy, W., & Wong, K. F. (2014). Ranking model selection and fusion for effective microblog search. In SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014 (pp. 21-26). Association for Computing Machinery, Inc. https://doi.org/10.1145/2632188.2632202

Ranking model selection and fusion for effective microblog search. / Wei, Zhongyu; Gao, Wei; El-Ganainy, Tarek; Magdy, Walid; Wong, Kam Fai.

SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014. Association for Computing Machinery, Inc, 2014. p. 21-26.

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

Wei, Z, Gao, W, El-Ganainy, T, Magdy, W & Wong, KF 2014, Ranking model selection and fusion for effective microblog search. in SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014. Association for Computing Machinery, Inc, pp. 21-26, 1st ACM International Workshop on Social Media Retrieval and Analysis, SoMeRA 2014 - Co-located with SIGIR 2014, Gold Coast, Australia, 11/7/14. https://doi.org/10.1145/2632188.2632202
Wei Z, Gao W, El-Ganainy T, Magdy W, Wong KF. Ranking model selection and fusion for effective microblog search. In SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014. Association for Computing Machinery, Inc. 2014. p. 21-26 https://doi.org/10.1145/2632188.2632202
Wei, Zhongyu ; Gao, Wei ; El-Ganainy, Tarek ; Magdy, Walid ; Wong, Kam Fai. / Ranking model selection and fusion for effective microblog search. SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014. Association for Computing Machinery, Inc, 2014. pp. 21-26
@inproceedings{de6762406f47486cb232b6c2ac4523c6,
title = "Ranking model selection and fusion for effective microblog search",
abstract = "Re-ranking was shown to have positive impact on the effectiveness for microblog search. Yet existing approaches mostly focused on using a single ranker to learn some better ranking function with respect to various relevance features. Given various available rank learners (such as learning to rank algorithms), in this work, we mainly study an orthogonal problem where multiple learned ranking models form an ensemble for re-ranking the retrieved tweets than just using a single ranking model in order to achieve higher search effectiveness. We explore the use of query-sensitive model selection and rank fusion methods based on the result lists produced from multiple rank learners. Base on the TREC microblog datasets, we found that our selection-based ensemble approach can significantly outperform using the single best ranker, and it also has clear advantage over the rank fusion that combines the results of all the available models.",
keywords = "Aggregation, Microblog search, Rank fusion, Ranker selection, Re-ranking, Twitter",
author = "Zhongyu Wei and Wei Gao and Tarek El-Ganainy and Walid Magdy and Wong, {Kam Fai}",
year = "2014",
month = "1",
day = "1",
doi = "10.1145/2632188.2632202",
language = "English",
isbn = "9781450330220",
pages = "21--26",
booktitle = "SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Ranking model selection and fusion for effective microblog search

AU - Wei, Zhongyu

AU - Gao, Wei

AU - El-Ganainy, Tarek

AU - Magdy, Walid

AU - Wong, Kam Fai

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Re-ranking was shown to have positive impact on the effectiveness for microblog search. Yet existing approaches mostly focused on using a single ranker to learn some better ranking function with respect to various relevance features. Given various available rank learners (such as learning to rank algorithms), in this work, we mainly study an orthogonal problem where multiple learned ranking models form an ensemble for re-ranking the retrieved tweets than just using a single ranking model in order to achieve higher search effectiveness. We explore the use of query-sensitive model selection and rank fusion methods based on the result lists produced from multiple rank learners. Base on the TREC microblog datasets, we found that our selection-based ensemble approach can significantly outperform using the single best ranker, and it also has clear advantage over the rank fusion that combines the results of all the available models.

AB - Re-ranking was shown to have positive impact on the effectiveness for microblog search. Yet existing approaches mostly focused on using a single ranker to learn some better ranking function with respect to various relevance features. Given various available rank learners (such as learning to rank algorithms), in this work, we mainly study an orthogonal problem where multiple learned ranking models form an ensemble for re-ranking the retrieved tweets than just using a single ranking model in order to achieve higher search effectiveness. We explore the use of query-sensitive model selection and rank fusion methods based on the result lists produced from multiple rank learners. Base on the TREC microblog datasets, we found that our selection-based ensemble approach can significantly outperform using the single best ranker, and it also has clear advantage over the rank fusion that combines the results of all the available models.

KW - Aggregation

KW - Microblog search

KW - Rank fusion

KW - Ranker selection

KW - Re-ranking

KW - Twitter

UR - http://www.scopus.com/inward/record.url?scp=84908657607&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84908657607&partnerID=8YFLogxK

U2 - 10.1145/2632188.2632202

DO - 10.1145/2632188.2632202

M3 - Conference contribution

SN - 9781450330220

SP - 21

EP - 26

BT - SoMeRA 2014 - Proceedings of the 1st ACM International Workshop on Social Media Retrieval and Analysis, Co-located with SIGIR 2014

PB - Association for Computing Machinery, Inc

ER -