de.mpg.escidoc.pubman.appbase.FacesBean
English
 
Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

Gaussian Process Classification for Segmenting and Annotating Sequences

MPS-Authors
http://pubman.mpdl.mpg.de/cone/persons/resource/persons83782

Altun,  Y
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons83975

Hofmann,  T
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available
Citation

Altun, Y., Hofmann, T., & Smola, A. (2004). Gaussian Process Classification for Segmenting and Annotating Sequences. In 21st International Conference on Machine Learning (ICML 2004) (pp. 25-32). New York, USA: ACM Press.


Cite as: http://hdl.handle.net/11858/00-001M-0000-0013-D8A5-0
Abstract
Many real-world classification tasks involve the prediction of multiple, inter-dependent class labels. A prototypical case of this sort deals with prediction of a sequence of labels for a sequence of observations. Such problems arise naturally in the context of annotating and segmenting observation sequences. This paper generalizes Gaussian Process classification to predict multiple labels by taking dependencies between neighboring labels into account. Our approach is motivated by the desire to retain rigorous probabilistic semantics, while overcoming limitations of parametric methods like Conditional Random Fields, which exhibit conceptual and computational difficulties in high-dimensional input spaces. Experiments on named entity recognition and pitch accent prediction tasks demonstrate the competitiveness of our approach.