TildeCRF: Conditional Random Fields for Logical Sequences

Abstract

Conditional Random Fields (CRFs) provide a powerful instrument for labeling sequences. So far, however, CRFs have only been considered for labeling sequences over flat alphabets. In this paper, we describe TildeCRF, the first method for training CRFs on logical sequences, i.e., sequences over an alphabet of logical atoms. TildeCRF’s key idea is to use relational regression trees in Dietterich et al.’s gradient tree boosting approach. Thus, the CRF potential functions are represented as weighted sums of relational regression trees. Experiments show a significant improvement over established results achieved with hidden Markov models and Fisher kernels for logical sequences.

Cite

Text

Gutmann and Kersting. "TildeCRF: Conditional Random Fields for Logical Sequences." European Conference on Machine Learning, 2006. doi:10.1007/11871842_20

Markdown

[Gutmann and Kersting. "TildeCRF: Conditional Random Fields for Logical Sequences." European Conference on Machine Learning, 2006.](https://mlanthology.org/ecmlpkdd/2006/gutmann2006ecml-tildecrf/) doi:10.1007/11871842_20

BibTeX

@inproceedings{gutmann2006ecml-tildecrf,
  title     = {{TildeCRF: Conditional Random Fields for Logical Sequences}},
  author    = {Gutmann, Bernd and Kersting, Kristian},
  booktitle = {European Conference on Machine Learning},
  year      = {2006},
  pages     = {174-185},
  doi       = {10.1007/11871842_20},
  url       = {https://mlanthology.org/ecmlpkdd/2006/gutmann2006ecml-tildecrf/}
}