Proceedings Article, Paper
@InProceedings
Beitrag in Tagungsband, Workshop


Show entries of:

this year (2019) | last year (2018) | two years ago (2017) | Notes URL

Action:

login to update

Options:








Author, Editor

Author(s):

Sizov, Sergej
Siersdorfer, Stefan
Weikum, Gerhard

dblp
dblp
dblp



Editor(s):

Evans, David A.
Gravano, Luis
Herzog, Otthein
Zhai, ChengXiang
Ronthaler, Marc

dblp
dblp
dblp
dblp
dblp

Not MPII Editor(s):

Evans, David A.
Gravano, Luis
Herzog, Otthein
Zhai, ChengXiang
Ronthaler, Marc

BibTeX cite key*:

SizovClass2004

Title, Booktitle

Title*:

Goal-oriented Methods and Meta Methods for Document Classification and their Parameter Tuning


cikm04-275-sizov.pdf (201.83 KB)

Booktitle*:

CIKM 2004 : proceedings of the Thirteenth Conference on Information and Knowledge Management

Event, URLs

URL of the conference:

http://ir.iit.edu/cikm2004/

URL for downloading the paper:


Event Address*:

Washington D.C., USA

Language:

English

Event Date*
(no longer used):


Organization:


Event Start Date:

8 November 2004

Event End Date:

13 November 2004

Publisher

Name*:

ACM

URL:


Address*:

New York, USA

Type:


Vol, No, Year, pp.

Series:


Volume:


Number:


Month:


Pages:

59-68

Year*:

2004

VG Wort Pages:


ISBN/ISSN:


Sequence Number:


DOI:




Note, Abstract, ©

Note:

Acceptance ratio 1:5

(LaTeX) Abstract:

Automatic text classification methods come with various
calibration parameters such as thresholds for probabilities in
Bayesian classifiers or for hyperplane distances in SVM
classifiers. In a given application context these parameters
should be set so as to meet the relative importance of various
result quality metrics such as precision versus recall. In this
paper we consider classifiers that can accept a document for a
topic, reject it, or abstain. We aim to meet the application's
goals in terms of accuracy (i.e., avoid false acceptances or
rejections) and loss (i.e., limit the fraction of documents for which no decision is
made).
To this end we investigate restrictive forms
of Support Vector Machine classifiers and we develop meta
methods that split the training data into subsets for
independently trained classifiers and then combine the results of
these classifiers. These techniques tend to improve accuracy at
the expense of document loss. We develop estimators that help to
predict the accuracy and loss for a given setting of the methods'
tuning parameters, and a methodology for efficiently deriving
a setting that meets the application's goals. Our experiments
confirm the practical viability of the approach.

Keywords:

Meta Classification, Restrictive Classification



Download
Access Level:

Public

Correlation

MPG Unit:

Max-Planck-Institut für Informatik



MPG Subunit:

Databases and Information Systems Group

Appearance:

MPII WWW Server, MPII FTP Server, MPG publications list, university publications list, working group publication list, Fachbeirat, VG Wort



BibTeX Entry:

@INPROCEEDINGS{SizovClass2004,
AUTHOR = {Sizov, Sergej and Siersdorfer, Stefan and Weikum, Gerhard},
EDITOR = {Evans, David A. and Gravano, Luis and Herzog, Otthein and Zhai, ChengXiang and Ronthaler, Marc},
TITLE = {Goal-oriented Methods and Meta Methods for Document Classification and their Parameter Tuning},
BOOKTITLE = {CIKM 2004 : proceedings of the Thirteenth Conference on Information and Knowledge Management},
PUBLISHER = {ACM},
YEAR = {2004},
PAGES = {59--68},
ADDRESS = {Washington D.C., USA},
NOTE = {Acceptance ratio 1:5},
}


Entry last modified by Christine Kiesel, 05/31/2005
Show details for Edit History (please click the blue arrow to see the details)Edit History (please click the blue arrow to see the details)
Hide details for Edit History (please click the blue arrow to see the details)Edit History (please click the blue arrow to see the details)

Editor(s)
Sergej Sizov
Created
01/26/2005 10:19:44 AM
Revisions
3.
2.
1.
0.
Editor(s)
Christine Kiesel
Petra Schaaf
Sabine Krott
Sergej Sizov
Edit Dates
31.05.2005 14:36:41
14.04.2005 10:30:09
09.02.2005 10:53:58
26.01.2005 10:19:45
Show details for Attachment SectionAttachment Section
Hide details for Attachment SectionAttachment Section

View attachments here:


File Attachment Icon
cikm04-275-sizov.pdf