Data fusion with estimated weight

S. Wu, F. Crestani

Research output: Contribution to conferencePaper

42 Citations (Scopus)

Abstract

This paper proposes an adptive approach for data fusion of information retrieval systems, which exploits estimated performances of all component input systems without relevance judgement or training. The estimation is conducted prior to the fusion but uses the same data as fusion applies. The experiment shows that our algorithms are competitive with, and often outperform CombMNZ, one of the most effective algorithms in use.

Conference

ConferenceProceedings of the eleventh international Conference on Information and Knowledge Management 2002
CityMcLean, USA
Period4/11/029/11/02

Fingerprint

Data fusion
Information retrieval systems
Experiments

Keywords

  • data fusion
  • distributed information retrieval
  • metasearch
  • results merging

Cite this

Wu, S., & Crestani, F. (2002). Data fusion with estimated weight. 648-651. Paper presented at Proceedings of the eleventh international Conference on Information and Knowledge Management 2002, McLean, USA, .
Wu, S. ; Crestani, F. / Data fusion with estimated weight. Paper presented at Proceedings of the eleventh international Conference on Information and Knowledge Management 2002, McLean, USA, .3 p.
@conference{4af569b1954a42c1ae68927f94c0e61b,
title = "Data fusion with estimated weight",
abstract = "This paper proposes an adptive approach for data fusion of information retrieval systems, which exploits estimated performances of all component input systems without relevance judgement or training. The estimation is conducted prior to the fusion but uses the same data as fusion applies. The experiment shows that our algorithms are competitive with, and often outperform CombMNZ, one of the most effective algorithms in use.",
keywords = "data fusion, distributed information retrieval, metasearch, results merging",
author = "S. Wu and F. Crestani",
year = "2002",
month = "11",
day = "5",
language = "English",
pages = "648--651",
note = "Proceedings of the eleventh international Conference on Information and Knowledge Management 2002 ; Conference date: 04-11-2002 Through 09-11-2002",

}

Wu, S & Crestani, F 2002, 'Data fusion with estimated weight' Paper presented at Proceedings of the eleventh international Conference on Information and Knowledge Management 2002, McLean, USA, 4/11/02 - 9/11/02, pp. 648-651.

Data fusion with estimated weight. / Wu, S.; Crestani, F.

2002. 648-651 Paper presented at Proceedings of the eleventh international Conference on Information and Knowledge Management 2002, McLean, USA, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Data fusion with estimated weight

AU - Wu, S.

AU - Crestani, F.

PY - 2002/11/5

Y1 - 2002/11/5

N2 - This paper proposes an adptive approach for data fusion of information retrieval systems, which exploits estimated performances of all component input systems without relevance judgement or training. The estimation is conducted prior to the fusion but uses the same data as fusion applies. The experiment shows that our algorithms are competitive with, and often outperform CombMNZ, one of the most effective algorithms in use.

AB - This paper proposes an adptive approach for data fusion of information retrieval systems, which exploits estimated performances of all component input systems without relevance judgement or training. The estimation is conducted prior to the fusion but uses the same data as fusion applies. The experiment shows that our algorithms are competitive with, and often outperform CombMNZ, one of the most effective algorithms in use.

KW - data fusion

KW - distributed information retrieval

KW - metasearch

KW - results merging

UR - http://dx.doi.org/10.1145/584792.584908

M3 - Paper

SP - 648

EP - 651

ER -

Wu S, Crestani F. Data fusion with estimated weight. 2002. Paper presented at Proceedings of the eleventh international Conference on Information and Knowledge Management 2002, McLean, USA, .