Deutsch Intern
    Data Science Chair


    Publications by Andreas Hotho

    These publications are hosted by BibSonomy.

    Stacked Conditional Random Fields Exploiting Structural Consistencies

    Klügl, Peter; Toepfer, Martin; Lemmerich, Florian; Hotho, Andreas; Puppe, Frank in Proceedings of 1st International Conference on Pattern Recognition Applications and Methods ICPRAM Carmona, Pedro Latorre; Sánchez, J. Salvador; Fred, Ana ( Eds. ), page 240-248 . Vilamoura, Algarve, Portugal , SciTePress , 2012 .

    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 structural consistencies. 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.
    Further Information
    Editor(s) Carmona, Pedro Latorre; Sánchez, J. Salvador; Fred, Ana
    Tags 2012  conditional  crf  fields  for:dmir  myown  random  stacked  sys:relevantfor:csuniwue  sys:relevantfor:l3s  sys:relevantfor:uniwue_info6