Beyond Relevance

Adapting Exploration/Exploitation in Information Retrieval

Kumaripaba Athukorala, Alan Medlar, Giulio Jacucci, Antti Oulasvirta, Dorota Glowacka

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

We present a novel adaptation technique for search engines to
better support information-seeking activities that include both
lookup and exploratory tasks. Building on previous findings,
we describe (1) a classifier that recognizes task type (lookup
vs. exploratory) as a user is searching and (2) a reinforcement
learning based search engine that adapts accordingly
the balance of exploration/exploitation in ranking documents.
This allows supporting both task types surreptitiously without
changing the familiar list-based interface. Search results include
more diverse results when users are exploring and more
precise results for lookup tasks. Users found more useful results
in exploratory tasks when compared to a baseline system,
which is specifically tuned for lookup tasks.
Original languageEnglish
Title of host publicationProceedings of the 21st International Conference on Intelligent User Interfaces
Number of pages10
Place of PublicationNew York, NY
PublisherACM, Association for Computing Machinery
Publication date7 Mar 2016
Pages359-369
ISBN (Print)978-1-4503-4137-0
DOIs
Publication statusPublished - 7 Mar 2016
MoE publication typeA4 Article in conference proceedings
EventInternational Conference on Intelligent User Interfaces - Sonoma, United States
Duration: 7 Mar 201610 Mar 2016
Conference number: 21 (IUI)

Bibliographical note

IUI'16
Volume:
Proceeding volume:

Fields of Science

  • 113 Computer and information sciences

Cite this

Athukorala, K., Medlar, A., Jacucci, G., Oulasvirta, A., & Glowacka, D. (2016). Beyond Relevance: Adapting Exploration/Exploitation in Information Retrieval. In Proceedings of the 21st International Conference on Intelligent User Interfaces (pp. 359-369). New York, NY: ACM, Association for Computing Machinery. https://doi.org/10.1145/2856767.2856786
Athukorala, Kumaripaba ; Medlar, Alan ; Jacucci, Giulio ; Oulasvirta, Antti ; Glowacka, Dorota. / Beyond Relevance : Adapting Exploration/Exploitation in Information Retrieval. Proceedings of the 21st International Conference on Intelligent User Interfaces. New York, NY : ACM, Association for Computing Machinery, 2016. pp. 359-369
@inproceedings{035f4ff7ebe24babb6e3affd22625f70,
title = "Beyond Relevance: Adapting Exploration/Exploitation in Information Retrieval",
abstract = "We present a novel adaptation technique for search engines tobetter support information-seeking activities that include bothlookup and exploratory tasks. Building on previous findings,we describe (1) a classifier that recognizes task type (lookupvs. exploratory) as a user is searching and (2) a reinforcementlearning based search engine that adapts accordinglythe balance of exploration/exploitation in ranking documents.This allows supporting both task types surreptitiously withoutchanging the familiar list-based interface. Search results includemore diverse results when users are exploring and moreprecise results for lookup tasks. Users found more useful resultsin exploratory tasks when compared to a baseline system,which is specifically tuned for lookup tasks.",
keywords = "113 Computer and information sciences",
author = "Kumaripaba Athukorala and Alan Medlar and Giulio Jacucci and Antti Oulasvirta and Dorota Glowacka",
note = "IUI'16 Volume: Proceeding volume:",
year = "2016",
month = "3",
day = "7",
doi = "10.1145/2856767.2856786",
language = "English",
isbn = "978-1-4503-4137-0",
pages = "359--369",
booktitle = "Proceedings of the 21st International Conference on Intelligent User Interfaces",
publisher = "ACM, Association for Computing Machinery",
address = "United States",

}

Athukorala, K, Medlar, A, Jacucci, G, Oulasvirta, A & Glowacka, D 2016, Beyond Relevance: Adapting Exploration/Exploitation in Information Retrieval. in Proceedings of the 21st International Conference on Intelligent User Interfaces. ACM, Association for Computing Machinery, New York, NY, pp. 359-369, International Conference on Intelligent User Interfaces, Sonoma, United States, 07/03/2016. https://doi.org/10.1145/2856767.2856786

Beyond Relevance : Adapting Exploration/Exploitation in Information Retrieval. / Athukorala, Kumaripaba; Medlar, Alan; Jacucci, Giulio; Oulasvirta, Antti; Glowacka, Dorota.

Proceedings of the 21st International Conference on Intelligent User Interfaces. New York, NY : ACM, Association for Computing Machinery, 2016. p. 359-369.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

TY - GEN

T1 - Beyond Relevance

T2 - Adapting Exploration/Exploitation in Information Retrieval

AU - Athukorala, Kumaripaba

AU - Medlar, Alan

AU - Jacucci, Giulio

AU - Oulasvirta, Antti

AU - Glowacka, Dorota

N1 - IUI'16 Volume: Proceeding volume:

PY - 2016/3/7

Y1 - 2016/3/7

N2 - We present a novel adaptation technique for search engines tobetter support information-seeking activities that include bothlookup and exploratory tasks. Building on previous findings,we describe (1) a classifier that recognizes task type (lookupvs. exploratory) as a user is searching and (2) a reinforcementlearning based search engine that adapts accordinglythe balance of exploration/exploitation in ranking documents.This allows supporting both task types surreptitiously withoutchanging the familiar list-based interface. Search results includemore diverse results when users are exploring and moreprecise results for lookup tasks. Users found more useful resultsin exploratory tasks when compared to a baseline system,which is specifically tuned for lookup tasks.

AB - We present a novel adaptation technique for search engines tobetter support information-seeking activities that include bothlookup and exploratory tasks. Building on previous findings,we describe (1) a classifier that recognizes task type (lookupvs. exploratory) as a user is searching and (2) a reinforcementlearning based search engine that adapts accordinglythe balance of exploration/exploitation in ranking documents.This allows supporting both task types surreptitiously withoutchanging the familiar list-based interface. Search results includemore diverse results when users are exploring and moreprecise results for lookup tasks. Users found more useful resultsin exploratory tasks when compared to a baseline system,which is specifically tuned for lookup tasks.

KW - 113 Computer and information sciences

U2 - 10.1145/2856767.2856786

DO - 10.1145/2856767.2856786

M3 - Conference contribution

SN - 978-1-4503-4137-0

SP - 359

EP - 369

BT - Proceedings of the 21st International Conference on Intelligent User Interfaces

PB - ACM, Association for Computing Machinery

CY - New York, NY

ER -

Athukorala K, Medlar A, Jacucci G, Oulasvirta A, Glowacka D. Beyond Relevance: Adapting Exploration/Exploitation in Information Retrieval. In Proceedings of the 21st International Conference on Intelligent User Interfaces. New York, NY: ACM, Association for Computing Machinery. 2016. p. 359-369 https://doi.org/10.1145/2856767.2856786