Jumat, 27 Juli 2012

Tutorial on how to use ACRF-GRMM tool

GRMM is a powerful tool to construct and perform inference on graphical models. This tutorial is just the extension version of tutorial given by Charles Sutton in http://mallet.cs.umass.edu/grmm/general_crfs.php for ACRF. ACRF is an interface given by GRMM to train CRF with arbitrary graphical structure (Skip-Chain CRF, Factorial CRF, Hierarchical CRF, or other Dynamic CRF models). Unfortunately, GRMM (ACRF as well) is lack of documentation and available tutorial on how to use it directly.
In this tutorial, we present simple experiments using Dynamic CRF (i.e., Factorial CRF) for Noun-Phrase Chunking and POS Tagging at the same time. Of course, we will use ACRF in this experiment. You can refer to a paper authored by Charles Sutton (Dynamic Conditional Random Fields: Factorized Probabilistic Models for Labeling and Segmenting Sequence Data, ICML 2004) before following this tutorial.

4 komentar:

  1. Hi Alfan,
    I was desperately looking for a tutorial on GRMM and came across your blog (I want to use it for some sequence tagging tasks related to dialog acts) but unfortunately the link to your tutorial is not working. Please see to it. I seriously hope it is just a temporary issue.


    1. Hi Rikku,
      sorry for late reply.

      you can try this link now : http://alfan.kaist.ac.kr/tutorial/grmm.htm

  2. Hi Alfan,

    I am Minsu, I found your good tutorial for ACRF. This is really great job. Where can I get the file http://alfan.kaist.ac.kr/homepage/resources/DCRF.java ?
    I think there is no link yet.

  3. Hi Alfan,

    I am a new user of GRMM as well, I wonder whether you could give me your email for some discussion with another kind of CRF grapht structure.