English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
  Graph Based Semi-Supervised Learning with Sharper Edges

Shin, H., Hill, N., & Rätsch, G. (2006). Graph Based Semi-Supervised Learning with Sharper Edges. In J. Fürnkranz, T. Scheffer, & M. Spiliopoulou (Eds.), Machine Learning: ECML 2006: 17th European Conference on Machine Learning Berlin, Germany, September 18-22, 2006 (pp. 401-412). Berlin, Germany: Springer.

Item is

Files

show Files

Locators

show
hide
Description:
-

Creators

show
hide
 Creators:
Shin, H1, Author              
Hill, NJ2, 3, Author              
Rätsch, G1, Author              
Affiliations:
1Friedrich Miescher Laboratory, Max Planck Society, Max-Planck-Ring 9, 72076 Tübingen, DE, ou_2575692              
2Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795              
3Max Planck Institute for Biological Cybernetics, Max Planck Society, Spemannstrasse 38, 72076 Tübingen, DE, ou_1497794              

Content

show
hide
Free keywords: -
 Abstract: In many graph-based semi-supervised learning algorithms, edge weights are assumed to be fixed and determined by the data pointsamp;amp;amp;amp;lsquo; (often symmetric)relationships in input space, without considering directionality. However, relationships may be more informative in one direction (e.g. from labelled to unlabelled) than in the reverse direction, and some relationships (e.g. strong weights between oppositely labelled points) are unhelpful in either direction. Undesirable edges may reduce the amount of influence an informative point can propagate to its neighbours -- the point and its outgoing edges have been ``blunted.amp;amp;amp;amp;lsquo;amp;amp;amp;amp;lsquo; We present an approach to ``sharpeningamp;amp;amp;amp;lsquo;amp;amp;amp;amp;lsquo; in which weights are adjusted to meet an optimization criterion wherever they are directed towards labelled points. This principle can be applied to a wide variety of algorithms. In the current paper, we present one ad hoc solution satisfying the principle, in order to show that it can improve performance on a number of publicly available benchmark data sets.

Details

show
hide
Language(s):
 Dates: 2006-09
 Publication Status: Published in print
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: DOI: 10.1007/11871842_39
BibTex Citekey: 4165
 Degree: -

Event

show
hide
Title: 17th European Conference on Machine Learning (ECML 2006)
Place of Event: Berlin, Germany
Start-/End Date: 2006-09-18 - 2006-09-22

Legal Case

show

Project information

show

Source 1

show
hide
Title: Machine Learning: ECML 2006: 17th European Conference on Machine Learning Berlin, Germany, September 18-22, 2006
Source Genre: Proceedings
 Creator(s):
Fürnkranz, J, Editor
Scheffer, T, Editor
Spiliopoulou, M, Editor
Affiliations:
-
Publ. Info: Berlin, Germany : Springer
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 401 - 412 Identifier: ISBN: 978-3-540-45375-8

Source 2

show
hide
Title: Lecture Notes in Computer Science
Source Genre: Series
 Creator(s):
Affiliations:
Publ. Info: -
Pages: - Volume / Issue: 4212 Sequence Number: - Start / End Page: - Identifier: -