Kernel Methods for Document Filtering
Kernel Methods for Document Filtering
This paper describes the algorithms implemented by the KerMIT consortium for its participation in the Trec 2002 Filtering track. The consortium submitted runs for the routing task using a linear SVM, for the batch task using the same SVM in combination with an innovation threshold-selection mechanism, and for the adaptive task using both a second-order perceptron and a combination of SVM and perceptron with uneven margin. Results seem to indicate that these algorithm performed relatively well on the extensive TREC benchmark.
computation learning theory, margin-based algorithms, information filtering, support vector, routing tasks, KerMIT, SVM
Shawe-Taylor, John
b1931d97-fdd0-4bc1-89bc-ec01648e928b
Cancedda, Nicola
02a07b95-589a-4d69-b505-73cd204c485f
Cesa-Bianchi, Nicolo
8da9ffa8-ed8c-4310-90a7-f38e593e94c6
Conconi, Alex
6e9df7c8-2905-4444-bab2-c45294582ef9
Gentile, Claudio
4511ebb9-4763-4b8c-af0a-f3df726cac56
Goutte, Cyril
28fb097a-3b48-4b1a-b223-9585dc422fdd
Graepel, Thore
f01fa538-c0f8-4e36-bbcc-698366e73f39
Li, Yaoyong
073211dd-f160-4e2b-b09a-a170d865140d
Renders, Jean-Michel
7e81f709-e29c-4803-bb3c-6dd43f1bbfd7
Voorhees, Ellen
57e0a2f2-d380-4b0a-99e8-394b60d9eb50
Buckland, Lori P
90555d7a-a52b-41e8-8683-4dc1de437f76
2002
Shawe-Taylor, John
b1931d97-fdd0-4bc1-89bc-ec01648e928b
Cancedda, Nicola
02a07b95-589a-4d69-b505-73cd204c485f
Cesa-Bianchi, Nicolo
8da9ffa8-ed8c-4310-90a7-f38e593e94c6
Conconi, Alex
6e9df7c8-2905-4444-bab2-c45294582ef9
Gentile, Claudio
4511ebb9-4763-4b8c-af0a-f3df726cac56
Goutte, Cyril
28fb097a-3b48-4b1a-b223-9585dc422fdd
Graepel, Thore
f01fa538-c0f8-4e36-bbcc-698366e73f39
Li, Yaoyong
073211dd-f160-4e2b-b09a-a170d865140d
Renders, Jean-Michel
7e81f709-e29c-4803-bb3c-6dd43f1bbfd7
Voorhees, Ellen
57e0a2f2-d380-4b0a-99e8-394b60d9eb50
Buckland, Lori P
90555d7a-a52b-41e8-8683-4dc1de437f76
Shawe-Taylor, John, Cancedda, Nicola, Cesa-Bianchi, Nicolo, Conconi, Alex, Gentile, Claudio, Goutte, Cyril, Graepel, Thore, Li, Yaoyong and Renders, Jean-Michel
(2002)
Kernel Methods for Document Filtering.
Voorhees, Ellen and Buckland, Lori P
(eds.)
The Eleventh Text Retrieval Conference (TREC 2002), Gaithersburg, Maryland, United States.
19 - 22 Nov 2002.
Record type:
Conference or Workshop Item
(Paper)
Abstract
This paper describes the algorithms implemented by the KerMIT consortium for its participation in the Trec 2002 Filtering track. The consortium submitted runs for the routing task using a linear SVM, for the batch task using the same SVM in combination with an innovation threshold-selection mechanism, and for the adaptive task using both a second-order perceptron and a combination of SVM and perceptron with uneven margin. Results seem to indicate that these algorithm performed relatively well on the extensive TREC benchmark.
Text
trec02_pro.pdf
- Other
More information
Published date: 2002
Additional Information:
Event Dates: 19 - 22 November 2002
Venue - Dates:
The Eleventh Text Retrieval Conference (TREC 2002), Gaithersburg, Maryland, United States, 2002-11-19 - 2002-11-22
Keywords:
computation learning theory, margin-based algorithms, information filtering, support vector, routing tasks, KerMIT, SVM
Organisations:
Electronics & Computer Science
Identifiers
Local EPrints ID: 261577
URI: http://eprints.soton.ac.uk/id/eprint/261577
PURE UUID: 6fb9bfba-0467-4462-9c60-83470f15fbb2
Catalogue record
Date deposited: 24 Nov 2005
Last modified: 14 Mar 2024 06:55
Export record
Contributors
Author:
John Shawe-Taylor
Author:
Nicola Cancedda
Author:
Nicolo Cesa-Bianchi
Author:
Alex Conconi
Author:
Claudio Gentile
Author:
Cyril Goutte
Author:
Thore Graepel
Author:
Yaoyong Li
Author:
Jean-Michel Renders
Editor:
Ellen Voorhees
Editor:
Lori P Buckland
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