Where academic tradition
meets the exciting future

Kernels Incorporating Word Positional Information in Natural Language Disambiguation Tasks

Tapio Pahikkala, Sampo Pyysalo, Filip Ginter, Jorma Boberg, Jouni Järvinen, Tapio Salakoski, Kernels Incorporating Word Positional Information in Natural Language Disambiguation Tasks. In: Ingrid Russell, Zdravko Markov (Eds.), Proceedings of the Eighteenth International Florida Artificial Intelligence Research Society Conference, 442-448, AAAI Press, 2005.

Abstract:

In this paper, we introduce a new kernel function designed for the problem of word sense disambiguation. The presented kernel function employs two different types of positional information related to words present in the contexts of the words to be disambiguated. For each pair of words in two contexts, the proposed kernel takes into account both their distances from the ambiguous words and also the difference of their mutual positions. We apply the kernel to contextsensitive spelling correction with SVMs and show that it significantly outperforms other considered kernels.

Files:

Abstract in PDF-format

BibTeX entry:

@INPROCEEDINGS{inpPaPyGiBoJaSa05a,
  title = {Kernels Incorporating Word Positional Information in Natural Language Disambiguation Tasks},
  booktitle = {Proceedings of the Eighteenth International Florida Artificial Intelligence Research Society Conference},
  author = {Pahikkala, Tapio and Pyysalo, Sampo and Ginter, Filip and Boberg, Jorma and Järvinen, Jouni and Salakoski, Tapio},
  editor = {Russell, Ingrid and Markov, Zdravko},
  publisher = {AAAI Press},
  pages = {442-448},
  year = {2005},
}

Belongs to TUCS Research Unit(s): Turku BioNLP Group

Edit publication