Probabilistic document-context based relevance feedback with limited relevance judgments

H. C. Wu; Wing Pong Robert Luk; K. F. Wong; K. L. Kwok

doi:10.1145/1183614.1183764

Probabilistic document-context based relevance feedback with limited relevance judgments

H. C. Wu, Wing Pong Robert Luk, K. F. Wong, K. L. Kwok

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

3 Citations (Scopus)

Abstract

This paper presents our novel relevance feedback (RF) algorithm that uses the probabilistic document-context based retrieval model with limited relevance judgments for document re-ranking. Probabilities of the document-context based retrieval model are estimated from the top N (=20) documents in the initial retrieval. We use document-context based cosine similarity measure to find similar data for better probability estimation in order to reduce the data scarcity problem and the negative weighting problem. Our RF algorithm is promising because its mean average precision is statistically significantly better than the baseline using TREC-6 and TREC-7 data collections.

Original language	English
Title of host publication	Proceedings of the 15th ACM Conference on Information and Knowledge Management, CIKM 2006
Pages	854-855
Number of pages	2
DOIs	https://doi.org/10.1145/1183614.1183764
Publication status	Published - 1 Dec 2006
Event	15th ACM Conference on Information and Knowledge Management, CIKM 2006 - Arlington, VA, United States Duration: 6 Nov 2006 → 11 Nov 2006

Conference

Conference	15th ACM Conference on Information and Knowledge Management, CIKM 2006
Country/Territory	United States
City	Arlington, VA
Period	6/11/06 → 11/11/06

Keywords

Document-context
Model

ASJC Scopus subject areas

General Decision Sciences
General Business,Management and Accounting

More information

10.1145/1183614.1183764

Cite this

@inproceedings{0da96f5416494135bc5efd027b29fedb,

title = "Probabilistic document-context based relevance feedback with limited relevance judgments",

abstract = "This paper presents our novel relevance feedback (RF) algorithm that uses the probabilistic document-context based retrieval model with limited relevance judgments for document re-ranking. Probabilities of the document-context based retrieval model are estimated from the top N (=20) documents in the initial retrieval. We use document-context based cosine similarity measure to find similar data for better probability estimation in order to reduce the data scarcity problem and the negative weighting problem. Our RF algorithm is promising because its mean average precision is statistically significantly better than the baseline using TREC-6 and TREC-7 data collections.",

keywords = "Document-context, Model",

author = "Wu, {H. C.} and Luk, {Wing Pong Robert} and Wong, {K. F.} and Kwok, {K. L.}",

year = "2006",

month = dec,

day = "1",

doi = "10.1145/1183614.1183764",

language = "English",

isbn = "1595934332",

pages = "854--855",

booktitle = "Proceedings of the 15th ACM Conference on Information and Knowledge Management, CIKM 2006",

note = "15th ACM Conference on Information and Knowledge Management, CIKM 2006 ; Conference date: 06-11-2006 Through 11-11-2006",

}

Wu, HC, Luk, WPR, Wong, KF & Kwok, KL 2006, Probabilistic document-context based relevance feedback with limited relevance judgments. in Proceedings of the 15th ACM Conference on Information and Knowledge Management, CIKM 2006. pp. 854-855, 15th ACM Conference on Information and Knowledge Management, CIKM 2006, Arlington, VA, United States, 6/11/06. https://doi.org/10.1145/1183614.1183764

Probabilistic document-context based relevance feedback with limited relevance judgments. / Wu, H. C.; Luk, Wing Pong Robert; Wong, K. F. et al.
Proceedings of the 15th ACM Conference on Information and Knowledge Management, CIKM 2006. 2006. p. 854-855.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Probabilistic document-context based relevance feedback with limited relevance judgments

AU - Wu, H. C.

AU - Luk, Wing Pong Robert

AU - Wong, K. F.

AU - Kwok, K. L.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - This paper presents our novel relevance feedback (RF) algorithm that uses the probabilistic document-context based retrieval model with limited relevance judgments for document re-ranking. Probabilities of the document-context based retrieval model are estimated from the top N (=20) documents in the initial retrieval. We use document-context based cosine similarity measure to find similar data for better probability estimation in order to reduce the data scarcity problem and the negative weighting problem. Our RF algorithm is promising because its mean average precision is statistically significantly better than the baseline using TREC-6 and TREC-7 data collections.

AB - This paper presents our novel relevance feedback (RF) algorithm that uses the probabilistic document-context based retrieval model with limited relevance judgments for document re-ranking. Probabilities of the document-context based retrieval model are estimated from the top N (=20) documents in the initial retrieval. We use document-context based cosine similarity measure to find similar data for better probability estimation in order to reduce the data scarcity problem and the negative weighting problem. Our RF algorithm is promising because its mean average precision is statistically significantly better than the baseline using TREC-6 and TREC-7 data collections.

KW - Document-context

KW - Model

UR - http://www.scopus.com/inward/record.url?scp=34547626063&partnerID=8YFLogxK

U2 - 10.1145/1183614.1183764

DO - 10.1145/1183614.1183764

M3 - Conference article published in proceeding or book

SN - 1595934332

SN - 9781595934338

SP - 854

EP - 855

BT - Proceedings of the 15th ACM Conference on Information and Knowledge Management, CIKM 2006

T2 - 15th ACM Conference on Information and Knowledge Management, CIKM 2006

Y2 - 6 November 2006 through 11 November 2006

ER -

Probabilistic document-context based relevance feedback with limited relevance judgments

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this