Difference between revisions of "Bellare 2009 generalized expectation criteria for bootstrapping extractors using record text alignment"

From Cohen Courses
Jump to navigationJump to search
Line 11: Line 11:
 
== Summary ==
 
== Summary ==
  
−
This [[Category::paper]] presents an [[UsesMethod::Active Learning]] approach that is not fully supervised. In this paper, the authors propose a semi-supervised approach where only some of the sequences are asked to be labeled. Assuming that there are subsequences that the model is confident about the labels even in a sequence that is uncertain as a whole, it only asks for labels for the subsequence the model is uncertain about and the rest is labeled using the current classifier. From their experiment this approach could save about 50~60% annotation labor over fully supervised active learning in the sequential learning settings.
+
This [[Category::paper]] presents a novel approach using [[UsesMethod::Generalized Expectation Criteria]] to train a [[UsesMethod::Conditional Random Field]] model for an IE task. In a setting where there exists a database
  
 
== Brief description of the method ==
 
== Brief description of the method ==
  
−
The method is a pretty simple extension of a standard active learning method. The following figure describes the general active learning framework.
+
The paper present two CRF models: AlignCRF and ExtrCRF. The first is a zero-order CRF model used to predict labels for a text sequence given a matching DB record. The other model is a first-order linear-chain CRF to extract when there is no DB record to match.  
  
−
[[File:Tomanek ACL2009.png]]
+
=== AlignCRF ===
  
−
The authors refer the usual active learning mode as fully supervised active learning (FuSAL). The utility function used in FuSAL is
 
  
−
<math>U_{\mathbf{\lambda}}(\mathbf{x}) = 1 - P_{\mathbf{\lambda}}(\mathbf{y}^{*}\vert\mathbf{x})</math>
 
  
−
which makes the sampling method as an uncertainty sampling method.
+
=== ExtrCRF ===
  
−
The problem of FuSAL in the sequence labeling scenario is that an example that has a high utility can still have parts of it that the current model can label very well, thus not contribute much to the utility of the whole. Therefore, it means we can leave some of the labels that the current model labeled if the confidence on that particular token is high enough. The authors name this as semi-supervised active learning (SeSAL). It combines the benefits of [[UsesMethod::Active Learning]] and [[UsesMethod::Bootstrapping]], which are labeling only examples with high utility and minimizing annotation effort by partially labeling examples where the model is confident about the prediction. In pseudocode, the following shows the steps that are added to the FuSAL:
+
=== Features ===
  
−
3.1 For each example <math>p_{i}\quad</math>
+
* Extraction features (in AlignCRF and ExtrCRF)
 +
** regular expressions detecting tokens containing all characters, all digits, or all alphanumeric
 +
** number of characters and digits in the token (ex. [NUMCHAR=3, NUMDIGITS=1])
 +
** domain-specific patterns for 'date', and 'pages'
 +
** token identity, prefix/suffix, character n-grams
 +
** presence of a token in lexicons such as "last names", "publisher names", "cities
 +
** lexicon features within a window of 10
 +
** regular expression feature within a window of 10
 +
** token identity features within a window of 3
  
−
3.1.1 For each token <math>x_{j}\quad</math> and the most likely label <math>y_{j}^{*}\quad</math>
+
* Alignment features (in AlignCRF)
−
 
+
** exact token match
−
3.1.1.1 Compute the model's confidence in the predicted label <math>C_{\mathbf{\lambda}}(y_{j}^{*})=P_{\mathbf{\lambda}}(y_{j}=y_{j}^{*}\vert\mathbf{x})</math>
+
** approximate token match after binning Jaro-Winkler edit distance between tokens
−
 
+
** substring token match
−
3.1.2 Remove all labels whose confidence is lower than some threshold <math>t</math>
+
** prefix/suffix token match (if the prefixes/suffixes match for lengths 1,2,3,4)
−
 
+
** exact and approximate token matches at offsets (-1,-1) and (+1,+1) around the alignment
−
Since there is a bootstrapping element in the method, the size of the seed set is also important. Therefore the authors suggest running FuSAL several iterations before switching to SeSAL.
 
  
 
== Experimental Result ==
 
== Experimental Result ==

Revision as of 07:41, 1 November 2011

A summary is coming soon from Daegunw!

Citation

Generalized Expectation Criteria for Bootstrapping Extractors using Record-Text Alignment, by K. Bellare, A. McCallum. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, 2009.

Online version

This Paper is available online [1].

Summary

This paper presents a novel approach using Generalized Expectation Criteria to train a Conditional Random Field model for an IE task. In a setting where there exists a database

Brief description of the method

The paper present two CRF models: AlignCRF and ExtrCRF. The first is a zero-order CRF model used to predict labels for a text sequence given a matching DB record. The other model is a first-order linear-chain CRF to extract when there is no DB record to match.

AlignCRF

ExtrCRF

Features

  • Extraction features (in AlignCRF and ExtrCRF)
    • regular expressions detecting tokens containing all characters, all digits, or all alphanumeric
    • number of characters and digits in the token (ex. [NUMCHAR=3, NUMDIGITS=1])
    • domain-specific patterns for 'date', and 'pages'
    • token identity, prefix/suffix, character n-grams
    • presence of a token in lexicons such as "last names", "publisher names", "cities
    • lexicon features within a window of 10
    • regular expression feature within a window of 10
    • token identity features within a window of 3
  • Alignment features (in AlignCRF)
    • exact token match
    • approximate token match after binning Jaro-Winkler edit distance between tokens
    • substring token match
    • prefix/suffix token match (if the prefixes/suffixes match for lengths 1,2,3,4)
    • exact and approximate token matches at offsets (-1,-1) and (+1,+1) around the alignment

Experimental Result

The authors tested this method on MUC-7 and the oncology part of PennBioIE corpus. The base learner used for the experiment is a linear-chain Conditional Random Fields. Features used are orthographical features (regexp patterns), lexical and morphological features (prefix, suffix, lemmatized tokens), and contextual features (features of neighbor tokens). In terms of the number of tokens that had to be labled to reach the maximal F-score, SeSAL could save about 60% over FuSAL, and 80% over random sampling. Having high confidence was also important because it could save the model from making errors in the early stages.

Related papers


Comment

If you're further interested in active learning for NLP, you might want to see Burr Settles' review of active learning: http://active-learning.net/ --Brendan 22:51, 13 October 2011 (UTC)

Retrieved from "http://curtis.ml.cmu.edu/w/courses/index.php?title=Bellare_2009_generalized_expectation_criteria_for_bootstrapping_extractors_using_record_text_alignment&oldid=9642"