Header logo is

Gaussian Process Classification for Segmenting and Annotating Sequences

2004

Conference Paper

ei


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.

Author(s): Altun, Y. and Hofmann, T. and Smola, AJ.
Journal: Proceedings of the 21st International Conference on Machine Learning (ICML 2004)
Pages: 25-32
Year: 2004
Month: July
Day: 0
Editors: Greiner, R. , D. Schuurmans
Publisher: ACM Press

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

DOI: 10.1145/1015330.1015433
Event Name: 21st International Conference on Machine Learning (ICML 2004)
Event Place: Banf, Alberta, Canada

Address: New York, USA
Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{2740,
  title = {Gaussian Process Classification for Segmenting and Annotating Sequences},
  author = {Altun, Y. and Hofmann, T. and Smola, AJ.},
  journal = {Proceedings of the 21st International Conference on Machine Learning (ICML 2004)},
  pages = {25-32},
  editors = {Greiner, R. , D. Schuurmans},
  publisher = {ACM Press},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {New York, USA},
  month = jul,
  year = {2004},
  doi = {10.1145/1015330.1015433},
  month_numeric = {7}
}