|
Published Articles >> Table of Contents >> Abstract
18th IEEE International Conference on Tools with Artificial Intelligence (ICTAI'06)
pp. 323-331
Multi-Criterion Active Learning in Conditional Random Fields
Christopher T. Symons, Oak Ridge National Laboratory, USA
Nagiza F. Samatova, Oak Ridge National Laboratory, USA
Ramya Krishnamurthy, Oak Ridge National Laboratory, USA
Byung H. Park, Oak Ridge National Laboratory, USA
Tarik Umar, Oak Ridge National Laboratory, USA
David Buttler, Lawrence Livermore National Laboratory, USA
Terence Critchlow, Lawrence Livermore National Laboratory, USA
David Hysom, Lawrence Livermore National Laboratory, USA
Full Article Text:
 
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICTAI.2006.90
Send link to a friend
| Abstract |
|
Conditional Random Fields (CRFs), which are
popular supervised learning models for many Natural
Language Processing (NLP) tasks, typically require a
large collection of labeled data for training. In
practice, however, manual annotation of text
documents is quite costly. Furthermore, even large
labeled training sets can have arbitrarily limited
performance peaks if they are not chosen with care.
This paper considers the use of multi-criterion active
learning for identification of a small but sufficient set
of text samples for training CRFs. Our empirical
results demonstrate that our method is capable of
reducing the manual annotation costs, while also
limiting the retraining costs that are often associated
with active learning. In addition, we show that the
generalization performance of CRFs can be enhanced
through judicious selection of training examples.
|
 Additional Information
|
Citation:
Christopher T. Symons, Nagiza F. Samatova, Ramya Krishnamurthy, Byung H. Park, Tarik Umar, David Buttler, Terence Critchlow, David Hysom,
"Multi-Criterion Active Learning in Conditional Random Fields,"
ictai,
pp. 323-331,
18th IEEE International Conference on Tools with Artificial Intelligence (ICTAI'06),
2006
|
|
