Ranking expansion terms using partial and ostensive evidence

Ian Ruthven; Mounia Lalmas; Keith van Rijsbergen; H. Bruce; R. Fidel; P. Ingwersen; P. Vakkari

Ranking expansion terms using partial and ostensive evidence

Ian Ruthven, Mounia Lalmas, Keith van Rijsbergen, H. Bruce (Editor), R. Fidel (Editor), P. Ingwersen (Editor), P. Vakkari (Editor)

Research output: Contribution to conference › Paper › peer-review

30 Downloads (Pure)

Abstract

In this paper we examine the problem of ranking candidate expansion terms for query expansion. We show, by an extension to the traditional F4 scheme, how partial relevance assessments (how relevant a document is) and ostensive evidence (when a document was assessed relevant) can be incorporated into a term ranking function. We then investigate this new term ranking function in three user experiments, examining the performance of our function for automatic and interactive query expansion. We show that the new function not only suggests terms that are preferred by searchers but suggests terms that can lead to more use of expansion terms.

Original language	English
Pages	199-220
Number of pages	21
Publication status	Published - 21 Jul 2002
Event	4th International Conference on Conceptions of Library and Information Science. CoLIS 4 - Seattle, USA Duration: 21 Jul 2002 → 25 Jul 2002

Conference

Conference	4th International Conference on Conceptions of Library and Information Science. CoLIS 4
City	Seattle, USA
Period	21/07/02 → 25/07/02

Keywords

expansion terms
query expansion
information retrieval
search algorithms

Access to Document

strathprints002462Accepted author manuscript, 273 KB

Cite this

@conference{cd548152170048e3a8c7bbee4287d415,

title = "Ranking expansion terms using partial and ostensive evidence",

abstract = "In this paper we examine the problem of ranking candidate expansion terms for query expansion. We show, by an extension to the traditional F4 scheme, how partial relevance assessments (how relevant a document is) and ostensive evidence (when a document was assessed relevant) can be incorporated into a term ranking function. We then investigate this new term ranking function in three user experiments, examining the performance of our function for automatic and interactive query expansion. We show that the new function not only suggests terms that are preferred by searchers but suggests terms that can lead to more use of expansion terms.",

keywords = "expansion terms, query expansion, information retrieval, search algorithms",

author = "Ian Ruthven and Mounia Lalmas and {van Rijsbergen}, Keith and H. Bruce and R. Fidel and P. Ingwersen and P. Vakkari",

year = "2002",

month = jul,

day = "21",

language = "English",

pages = "199--220",

note = "4th International Conference on Conceptions of Library and Information Science. CoLIS 4 ; Conference date: 21-07-2002 Through 25-07-2002",

}

TY - CONF

T1 - Ranking expansion terms using partial and ostensive evidence

AU - Ruthven, Ian

AU - Lalmas, Mounia

AU - van Rijsbergen, Keith

A2 - Bruce, H.

A2 - Fidel, R.

A2 - Ingwersen, P.

A2 - Vakkari, P.

PY - 2002/7/21

Y1 - 2002/7/21

N2 - In this paper we examine the problem of ranking candidate expansion terms for query expansion. We show, by an extension to the traditional F4 scheme, how partial relevance assessments (how relevant a document is) and ostensive evidence (when a document was assessed relevant) can be incorporated into a term ranking function. We then investigate this new term ranking function in three user experiments, examining the performance of our function for automatic and interactive query expansion. We show that the new function not only suggests terms that are preferred by searchers but suggests terms that can lead to more use of expansion terms.

AB - In this paper we examine the problem of ranking candidate expansion terms for query expansion. We show, by an extension to the traditional F4 scheme, how partial relevance assessments (how relevant a document is) and ostensive evidence (when a document was assessed relevant) can be incorporated into a term ranking function. We then investigate this new term ranking function in three user experiments, examining the performance of our function for automatic and interactive query expansion. We show that the new function not only suggests terms that are preferred by searchers but suggests terms that can lead to more use of expansion terms.

KW - expansion terms

KW - query expansion

KW - information retrieval

KW - search algorithms

UR - http://www.worldcat.org/oclc/224076313

M3 - Paper

SP - 199

EP - 220

T2 - 4th International Conference on Conceptions of Library and Information Science. CoLIS 4

Y2 - 21 July 2002 through 25 July 2002

ER -

Ranking expansion terms using partial and ostensive evidence

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this