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

From Cohen Courses
Jump to navigationJump to search
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
A summary is coming soon from [[User::Daegunw]]!
 
 
 
== Citation ==
 
== Citation ==
  
Line 11: Line 9:
 
== Summary ==
 
== Summary ==
  
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  
+
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, the authors train a CRF model for alignment with the unlabeled text using generalized expectation. Also, based on the alignment model, they also propose a usual 1st order CRF model that can extract information without relying on a DB record.
  
 
== Brief description of the method ==
 
== Brief description of the method ==
Line 38: Line 36:
 
=== AlignCRF ===
 
=== AlignCRF ===
  
For expectation features, for each label, the top N extraction features are selected by mutual information with that label. Also, the top N alignment features that have highest mutual information with correct labeling are selected as alignment criteria. The target expectations of these criteria are binned into 11 bins [0.05, 0.1, 0.2, ..., 0.9, 0.95]. [[http://www.cs.umass.edu/~kedarb/dbie_expts.txt]] is a complete list of expectation criteria.
+
For a database record with token sequence and label sequence <math>(\mathbf{x}_{1}, \mathbf{y}_{1})=(<{x}_{1}[1], {x}_{1}[2], ..., {x}_{1}[m]>, <{y}_{1}[1], ..., {y}_{1}[m]>)</math>, a text sequence <math>\mathbf{x}_{2}=<{x}_{2}[1], {x}_{2}[2], ..., {x}_{2}[m]></math> and an alignment sequence <math>\mathbf{a}=<a_1, ..., a_n></math> where <math> \quad a_{i}=j </math> indicates <math>\quad(x_1[j], y_1[j]) </math> is assigned to the token <math>\quad x_{2}[i]</math>, the alignment model defines a probability distribution of the alignment sequence conditioned on the database record and the text sequence <math>(\mathbf{x}_{1}, \mathbf{y}_{1}, \mathbf{x}_{2})</math> as
 +
 
 +
<math>
 +
p_{\Theta}(\mathbf{a}\vert\mathbf{x}_{1}, \mathbf{y}_{1}, \mathbf{x}_{2};\Theta)
 +
={exp(\sum_{t=1}^{n}{\Theta^{\mathrm{T}}\vec{f}(a',\mathbf{x}_1,\mathbf{y}_1,\mathbf{x}_2,t)})\over Z_{\Theta}(\mathbf{x}_1,\mathbf{y}_1,\mathbf{x}_2)}
 +
</math>
  
For a database record with token sequence and label sequence <math>(\mathbf{x}_{1}, \mathbf{y}_{1})=(<{x}_{1}[1], {x}_{1}[2], ..., {x}_{1}[m]>, <{y}_{1}[1], ..., {y}_{1}[m]>)</math>, a text sequence <math>\mathbf{x}_{2}=<{x}_{2}[1], {x}_{2}[2], ..., {x}_{2}[m]></math> and an alignment sequence <math>\mathbf{a}=<a_1, ..., a_n></math> where <math> a_{i}=j </math> indicates <math>(x_1[j], y_1[j]) </math> is assigned to the token <math>x_{2}[i]</math>, the alignment model defines a probability distribution over
+
<math>
 +
p_{\Theta}(a_{t}\vert\mathbf{x}_{1}, \mathbf{y}_{1}, \mathbf{x}_{2};\Theta)
 +
={exp(\Theta^{\mathrm{T}}\vec{f}(a_t,\mathbf{x}_1,\mathbf{y}_1,\mathbf{x}_2,t))\over
 +
exp(\sum_{a'}{\Theta^{\mathrm{T}}\vec{f}(a_t,\mathbf{x}_1,\mathbf{y}_1,\mathbf{x}_2,t)})}
 +
</math>
 +
 
 +
<math>
 +
{\partial p_{\Theta}(a_t\vert\mathbf{x}_{1}, \mathbf{y}_{1}, \mathbf{x}_{2};\Theta) \over \partial \Theta}
 +
=p_{\Theta}(a_{t}\vert\mathbf{x}_{1}, \mathbf{y}_{1}, \mathbf{x}_{2};\Theta)
 +
[
 +
\vec{f}(a_t,\mathbf{x}_1,\mathbf{y}_1,\mathbf{x}_2,t)
 +
-E_{p_{\Theta}(a)}(\vec{f}(a,\mathbf{x}_1,\mathbf{y}_1,\mathbf{x}_2,t))
 +
]
 +
</math>
 +
 
 +
[[File:Bellare 1.png]]
 +
 
 +
The expectation criteria is determined in the following manner: for expectation features, for each label, the top N extraction features are selected by mutual information with that label. Also, the top N alignment features that have highest mutual information with correct labeling are selected as alignment criteria. The target expectations of these criteria are binned into 11 bins [0.05, 0.1, 0.2, ..., 0.9, 0.95]. [[http://www.cs.umass.edu/~kedarb/dbie_expts.txt]] is a complete list of expectation criteria.
  
 
Given expectation criteria  <math>\mathcal{C} = <\mathbf{F}, \mathbf{P}, \mathbf{W}></math> where <math>\mathbf{F}=<f_1,...,f_l></math> is a list of binary feature functions, <math>\mathbf{P}=<p_1,...,p_l></math> is target expectations, <math>\mathbf{W}=<w_1,...,w_l></math> is weights, the objective function to optimize is the following:
 
Given expectation criteria  <math>\mathcal{C} = <\mathbf{F}, \mathbf{P}, \mathbf{W}></math> where <math>\mathbf{F}=<f_1,...,f_l></math> is a list of binary feature functions, <math>\mathbf{P}=<p_1,...,p_l></math> is target expectations, <math>\mathbf{W}=<w_1,...,w_l></math> is weights, the objective function to optimize is the following:
Line 52: Line 72:
  
 
where <math>
 
where <math>
\Delta(f_{i}, p_{i}, w_{i}, \Theta) = w( {E_{p_{\Theta}}(\mathbf{A}_{f}) \over |\mathbf{A}_{f}|} - p)^{2}
+
\Delta(f, p, w, \Theta) = w( {E_{p_{\Theta}}(\mathbf{A}_{f}) \over |\mathbf{A}_{f}|} - p)^{2}
</math> is  
+
</math> is the squared divergence and
 
 
 
<math>
 
<math>
{\partial\mathcal{O}(\theta;\mathcal{D},\mathcal{C}) \over \partial\Theta}=
+
\mathbf{A}_{f}  
  -\sum_{i=1}^{l}{\partial\Delta(f_{i}, p_{i}, w_{i}, \Theta)\over\partial\Theta} - \Theta
+
</math> is the alignment latent variables to apply expectation criteria. The authors use the L-BFGS algorithm to maximize the objective function.
</math>
 
  
 +
=== ExtrCRF ===
  
 +
This model, whose probability is denoted by <math>p_{\Lambda}</math>, is trained by minimizing the following objective function using L-BFGS.
  
=== ExtrCRF ===
+
<math>
 +
O(\Lambda;\mathcal{D},\Theta)=\min_{\Lambda}{\sum_{i=1}^{K}{KL(p_{\Theta}(\mathbf{y}\vert\mathbf{X}_{2}^{(i)})\vert\vert p_{\Lambda}(\mathbf{y}\vert\mathbf{X}_{2}^{(i)}))}+||\Lambda||^{2}/2}
 +
</math>
  
 
== Experimental Result ==
 
== Experimental Result ==
  
The authors tested this method on [[UsesDataset::MUC]]-7 and the oncology part of [[UsesDataset::PennBioIE]] corpus. The base learner used for the experiment is a linear-chain [[UsesMethod::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.
+
The authors tested this method on [[UsesDataset::DBLP]] bibliographic database on a pruned label set (author, title, date, venue, volume, number, pages, editor, publisher, series, O). AlignCRF clearly outperforms other models, and ExtrCRF also achieves an error reduction of 20~35% compared to other methods. An interesting thing to note is that there is no evident decrease in the performance of AlignCRF compared to ExtrCRF, although AlignCRF is not using DB records. This is due to the benefit of having higher order (first-order) model and using noisy DB records in the test set for alignment.
 +
 
 +
[[File:Bellare 2.png]]
 +
[[File:Bellare 3.png]]
  
 
== Related papers ==
 
== Related papers ==
  
* [[RelatedPaper::Muslea, Minton and Knoblock, ICML 2002]]
+
* [[RelatedPaper::Sarawagi_and_Cohen_NIPS_2004]]
* [[RelatedPaper::McCallum and Ngiam, ICML 98]]
+
* [[RelatedPaper::Mann and McCallum, ACL 08]]
 
 
 
 
== 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/  --[[User:Brendan|Brendan]] 22:51, 13 October 2011 (UTC)
 

Latest revision as of 12:09, 2 November 2011

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, the authors train a CRF model for alignment with the unlabeled text using generalized expectation. Also, based on the alignment model, they also propose a usual 1st order CRF model that can extract information without relying on a DB record.

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.

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

AlignCRF

For a database record with token sequence and label sequence , a text sequence and an alignment sequence where indicates is assigned to the token , the alignment model defines a probability distribution of the alignment sequence conditioned on the database record and the text sequence as

Bellare 1.png

The expectation criteria is determined in the following manner: for expectation features, for each label, the top N extraction features are selected by mutual information with that label. Also, the top N alignment features that have highest mutual information with correct labeling are selected as alignment criteria. The target expectations of these criteria are binned into 11 bins [0.05, 0.1, 0.2, ..., 0.9, 0.95]. [[2]] is a complete list of expectation criteria.

Given expectation criteria where is a list of binary feature functions, is target expectations, is weights, the objective function to optimize is the following:

where is the squared divergence and is the alignment latent variables to apply expectation criteria. The authors use the L-BFGS algorithm to maximize the objective function.

ExtrCRF

This model, whose probability is denoted by , is trained by minimizing the following objective function using L-BFGS.

Experimental Result

The authors tested this method on DBLP bibliographic database on a pruned label set (author, title, date, venue, volume, number, pages, editor, publisher, series, O). AlignCRF clearly outperforms other models, and ExtrCRF also achieves an error reduction of 20~35% compared to other methods. An interesting thing to note is that there is no evident decrease in the performance of AlignCRF compared to ExtrCRF, although AlignCRF is not using DB records. This is due to the benefit of having higher order (first-order) model and using noisy DB records in the test set for alignment.

Bellare 2.png Bellare 3.png

Related papers