| Supplement of partial ranks to the data fusion |
| Full text |
Pdf
(451 KB)
|
| Source
|
ACM International Conference Proceeding Series; Vol. 192
archive
Proceedings of the 12th Brazilian symposium on Multimedia and the web
table of contents
Natal, Rio Grande do Norte, Brazil
SESSION: Full papers (written in Portuguese)
table of contents
Pages: 148 - 154
Year of Publication: 2006
ISBN:85-7669-100-0
|
|
Authors
|
|
| Sponsor |
|
| Publisher |
|
| Bibliometrics |
Downloads (6 Weeks): 0, Downloads (12 Months): 19, Citation Count: 0
|
|
|
ABSTRACT
In this paper a model for the data fusion is proposed in what the ranks involved can differ in size and in elements. The fusion process allows establishing different weights for each involved rank, making possible to establish levels of importance according to the necessity of each fusion. The model also allows establishing truncate value, with the purpose to relate only the first top k fusion elements. It is also added to this paper, the ranks' supplement definition, with the purpose to consider in a rank the others contributions, whose elements are not present in the first. Experiments with many kinds of ranks are presented, with the purpose to analyze the fusion behavior with the ranks' supplement. With them it is possible to verify that the supplement does not affect the fusion ranks quality, bringing the elements to the top list.Neste artigo, propõe-se um modelo para a fusão de dados no qual os ranks envolvidos podem diferir em tamanho e elementos. O processo de fusão permite estabelecer pesos diferenciados para cada rank envolvido, possibilitando estabelecer níveis de relevância de acordo com a necessidade de cada fusão. O modelo ainda permite que se estabeleça um valor de truncamento, com a finalidade de relacionar apenas os primeiros k elementos da fusão. É introduzido, neste trabalho, a definição de completamento de ranks, com a finalidade de considerar em um rank as contribuições de outro, cujos elementos não estão presentes no primeiro. São apresentados experimentos com vários tipos de ranks, com o objetivo de analisar o comportamento da fusão com o completamento de ranks. Com estes é possível verificar que o completamento não afeta a qualidade da fusão dos ranks, retornando os elementos relevantes no topo da lista.
REFERENCES
Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.
 |
1
|
Cynthia Dwork , Ravi Kumar , Moni Naor , D. Sivakumar, Rank aggregation methods for the Web, Proceedings of the 10th international conference on World Wide Web, p.613-622, May 01-05, 2001, Hong Kong, Hong Kong
[doi> 10.1145/371920.372165]
|
 |
2
|
Ronald Fagin , Ravi Kumar , Mohammad Mahdian , D. Sivakumar , Erik Vee, Comparing and aggregating rankings with ties, Proceedings of the twenty-third ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems, June 14-16, 2004, Paris, France
[doi> 10.1145/1055558.1055568]
|
| |
3
|
Hsu, F., and Palumbo, A. A study of data fusion in Cayley Graphs G(Sn, Pn). In Proceedings of the 7th Intern. Symposium on Parallel Architectures, Algorithms and Networks (ISPAN'04) (May 10-12, 2004) IEEE Computer Society Press, 557--562.
|
| |
4
|
|
| |
5
|
Marden, J. I. Analyzing and Modeling Data Rank. Monographs on Statistics and Applied Probability, No 64. Chapman & Hall, 1995.
|
| |
6
|
|
|