ML Anthology

Revisiting Fisher Kernels for Document Similarities

Martin Nyffenegger, Jean-Cédric Chappelier, Éric Gaussier
ECML-PKDD 2006 pp. 727-734
doi:10.1007/11871842_73 /ecmlpkdd/2006/nyffenegger2006ecml-revisiting/

Abstract

This paper presents a new metric to compute similarities between textual documents, based on the Fisher information kernel as proposed by T. Hofmann. By considering a new point-of-view on the embedding vector space and proposing a more appropriate way of handling the Fisher information matrix, we derive a new form of the kernel that yields significant improvements on an information retrieval task. We apply our approach to two different models: Naive Bayes and PLSI.

PDF ECML-PKDD Semantic Scholar

Cite

Text

Nyffenegger et al. "Revisiting Fisher Kernels for Document Similarities." European Conference on Machine Learning, 2006. doi:10.1007/11871842_73

Markdown

[Nyffenegger et al. "Revisiting Fisher Kernels for Document Similarities." European Conference on Machine Learning, 2006.](https://mlanthology.org/ecmlpkdd/2006/nyffenegger2006ecml-revisiting/) doi:10.1007/11871842_73

BibTeX

@inproceedings{nyffenegger2006ecml-revisiting,
  title     = {{Revisiting Fisher Kernels for Document Similarities}},
  author    = {Nyffenegger, Martin and Chappelier, Jean-Cédric and Gaussier, Éric},
  booktitle = {European Conference on Machine Learning},
  year      = {2006},
  pages     = {727-734},
  doi       = {10.1007/11871842_73},
  url       = {https://mlanthology.org/ecmlpkdd/2006/nyffenegger2006ecml-revisiting/}
}