Transduction with Confidence and Credibility
Transduction with Confidence and Credibility
In this paper we follow the same general ideology as in (Gammerman et. al, 1998), and describe a new transductive learning algorithm using Support Vector Machines. The algorithm presented provides confidence values for its predicted classifications of new examples. We also obtain a measure of "credibility" which serves as an indicator of the reliability of the data upon which we make our prediction. Experiments compare the new algorithm to a standard Support Vector Machine and other transductive methods which use Support Vector Machines, such as Vapnik's margin transduction. Empirical results show that the new algorithm not only produces confidence and credibility measures, but is comparable to, and sometimes exceeds the performance of the other algorithms.
722-726
Saunders, C.
38a38da8-1eb3-47a8-80bc-b9cbb43f26e3
Gammerman, A.
b315c69d-8ac1-41c4-9617-3cccb95384aa
Vovk, V.
1feb1a01-8acd-4af5-9832-942537c296ed
1999
Saunders, C.
38a38da8-1eb3-47a8-80bc-b9cbb43f26e3
Gammerman, A.
b315c69d-8ac1-41c4-9617-3cccb95384aa
Vovk, V.
1feb1a01-8acd-4af5-9832-942537c296ed
Saunders, C., Gammerman, A. and Vovk, V.
(1999)
Transduction with Confidence and Credibility.
Sixteenth International Joint Conference on Artificial Intelligence (IJCAI '99).
.
Record type:
Conference or Workshop Item
(Other)
Abstract
In this paper we follow the same general ideology as in (Gammerman et. al, 1998), and describe a new transductive learning algorithm using Support Vector Machines. The algorithm presented provides confidence values for its predicted classifications of new examples. We also obtain a measure of "credibility" which serves as an indicator of the reliability of the data upon which we make our prediction. Experiments compare the new algorithm to a standard Support Vector Machine and other transductive methods which use Support Vector Machines, such as Vapnik's margin transduction. Empirical results show that the new algorithm not only produces confidence and credibility measures, but is comparable to, and sometimes exceeds the performance of the other algorithms.
Text
Transduction_IJCAI99.pdf
- Other
More information
Published date: 1999
Venue - Dates:
Sixteenth International Joint Conference on Artificial Intelligence (IJCAI '99), 1999-01-01
Organisations:
Electronics & Computer Science
Identifiers
Local EPrints ID: 258961
URI: http://eprints.soton.ac.uk/id/eprint/258961
PURE UUID: 4e7bcf59-1cd3-458c-8e48-7fb735a39465
Catalogue record
Date deposited: 03 Mar 2004
Last modified: 14 Mar 2024 06:16
Export record
Contributors
Author:
C. Saunders
Author:
A. Gammerman
Author:
V. Vovk
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics