Deriving implicit user feedback from partial URLs for effective web page retrieval

Rongmei Li; Theo van der Weide

Deriving implicit user feedback from partial URLs for effective web page retrieval

Rongmei Li, Theo van der Weide

Databases (Former)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

Abstract

User click-throughs provide a search context for understanding the user need of complex information. This paper re-examines the effectiveness of this approach when based on partial clicked data using the language modeling framework. We expand the original query by topical terms derived from clicked Web pages and enhance early precision via a more compact document representation. Since our URLs of Web pages are stripped, we first reconstruct them at different levels based on different collections. Our experimental results on the GOV2 test collection and AOL query log show improvement by 31.7% and 28.3% significantly in statMAP for two sources of reconstruction and 153 ad-hoc queries. Our model also outperforms pseudo relevance feedback.

Original language	English
Title of host publication	RIAO '10
Subtitle of host publication	Adaptivity, Personalization and Fusion of Heterogeneous Information
Publisher	ACM Press
Pages	124-125
Publication status	Published - 2010
Event	9th International Conference on Computer-Assisted Information Retrieval, RIAO 2010: Adaptivity, Personalization and Fusion of Heterogeneous Information (Recherche d'Information et ses Applications) - Bibliotheque Nationale de France, Paris, France Duration: 28 Apr 2010 → 30 Apr 2010 Conference number: 9

Conference

Conference	9th International Conference on Computer-Assisted Information Retrieval, RIAO 2010
Abbreviated title	RIAO
Country	France
City	Paris
Period	28/04/10 → 30/04/10

Keywords

IR-78541

Access to Document

https://dl.acm.org/citation.cfm?id=1937055.1937083

Fingerprint Dive into the research topics of 'Deriving implicit user feedback from partial URLs for effective web page retrieval'. Together they form a unique fingerprint.

Cite this

@inproceedings{6a3bed2f74724762bc6400e1c4c67abc,

title = "Deriving implicit user feedback from partial URLs for effective web page retrieval",

abstract = "User click-throughs provide a search context for understanding the user need of complex information. This paper re-examines the effectiveness of this approach when based on partial clicked data using the language modeling framework. We expand the original query by topical terms derived from clicked Web pages and enhance early precision via a more compact document representation. Since our URLs of Web pages are stripped, we first reconstruct them at different levels based on different collections. Our experimental results on the GOV2 test collection and AOL query log show improvement by 31.7% and 28.3% significantly in statMAP for two sources of reconstruction and 153 ad-hoc queries. Our model also outperforms pseudo relevance feedback.",

keywords = "IR-78541",

author = "Rongmei Li and {van der Weide}, Theo",

year = "2010",

language = "English",

pages = "124--125",

booktitle = "RIAO '10",

publisher = "ACM Press",

note = "9th International Conference on Computer-Assisted Information Retrieval, RIAO 2010 : Adaptivity, Personalization and Fusion of Heterogeneous Information (Recherche d'Information et ses Applications) , RIAO ; Conference date: 28-04-2010 Through 30-04-2010",

}

Li, R & van der Weide, T 2010, Deriving implicit user feedback from partial URLs for effective web page retrieval. in RIAO '10: Adaptivity, Personalization and Fusion of Heterogeneous Information. ACM Press, pp. 124-125, 9th International Conference on Computer-Assisted Information Retrieval, RIAO 2010, Paris, France, 28/04/10. <https://dl.acm.org/citation.cfm?id=1937055.1937083>

TY - GEN

T1 - Deriving implicit user feedback from partial URLs for effective web page retrieval

AU - Li, Rongmei

AU - van der Weide, Theo

N1 - Conference code: 9

PY - 2010

Y1 - 2010

N2 - User click-throughs provide a search context for understanding the user need of complex information. This paper re-examines the effectiveness of this approach when based on partial clicked data using the language modeling framework. We expand the original query by topical terms derived from clicked Web pages and enhance early precision via a more compact document representation. Since our URLs of Web pages are stripped, we first reconstruct them at different levels based on different collections. Our experimental results on the GOV2 test collection and AOL query log show improvement by 31.7% and 28.3% significantly in statMAP for two sources of reconstruction and 153 ad-hoc queries. Our model also outperforms pseudo relevance feedback.

AB - User click-throughs provide a search context for understanding the user need of complex information. This paper re-examines the effectiveness of this approach when based on partial clicked data using the language modeling framework. We expand the original query by topical terms derived from clicked Web pages and enhance early precision via a more compact document representation. Since our URLs of Web pages are stripped, we first reconstruct them at different levels based on different collections. Our experimental results on the GOV2 test collection and AOL query log show improvement by 31.7% and 28.3% significantly in statMAP for two sources of reconstruction and 153 ad-hoc queries. Our model also outperforms pseudo relevance feedback.

KW - IR-78541

M3 - Conference contribution

SP - 124

EP - 125

BT - RIAO '10

PB - ACM Press

T2 - 9th International Conference on Computer-Assisted Information Retrieval, RIAO 2010

Y2 - 28 April 2010 through 30 April 2010

ER -