STACKED CONDITIONAL RANDOM FIELDS EXPLOITING STRUCTURAL CONSISTENCIES
Peter Kluegl, Martin Toepfer, Florian Lemmerich, Andreas Hotho, Frank Puppe
2012
Abstract
Conditional Random Fields (CRF) are popular methods for labeling unstructured or textual data. Like many machine learning approaches these undirected graphical models assume the instances to be independently distributed. However, in real world applications data is grouped in a natural way, e.g., by its creation context. The instances in each group often share additional consistencies in the structure of their information. This paper proposes a domain-independent method for exploiting these consistencies by combining two CRFs in a stacked learning framework. The approach incorporates three successive steps of inference: First, an initial CRF processes single instances as usual. Next, we apply rule learning collectively on all labeled outputs of one context to acquire descriptions of its specific properties. Finally, we utilize these descriptions as dynamic and high quality features in an additional (stacked) CRF. The presented approach is evaluated with a real-world dataset for the segmentation of references and achieves a significant reduction of the labeling error.
DownloadPaper Citation
in Harvard Style
Kluegl P., Toepfer M., Lemmerich F., Hotho A. and Puppe F. (2012). STACKED CONDITIONAL RANDOM FIELDS EXPLOITING STRUCTURAL CONSISTENCIES . In Proceedings of the 1st International Conference on Pattern Recognition Applications and Methods - Volume 2: ICPRAM, ISBN 978-989-8425-99-7, pages 240-249. DOI: 10.5220/0003706602400249
in Bibtex Style
@conference{icpram12,
author={Peter Kluegl and Martin Toepfer and Florian Lemmerich and Andreas Hotho and Frank Puppe},
title={STACKED CONDITIONAL RANDOM FIELDS EXPLOITING STRUCTURAL CONSISTENCIES},
booktitle={Proceedings of the 1st International Conference on Pattern Recognition Applications and Methods - Volume 2: ICPRAM,},
year={2012},
pages={240-249},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003706602400249},
isbn={978-989-8425-99-7},
}
in EndNote Style
TY - CONF
JO - Proceedings of the 1st International Conference on Pattern Recognition Applications and Methods - Volume 2: ICPRAM,
TI - STACKED CONDITIONAL RANDOM FIELDS EXPLOITING STRUCTURAL CONSISTENCIES
SN - 978-989-8425-99-7
AU - Kluegl P.
AU - Toepfer M.
AU - Lemmerich F.
AU - Hotho A.
AU - Puppe F.
PY - 2012
SP - 240
EP - 249
DO - 10.5220/0003706602400249