Page History

{scrollbar}

...

line 2: <?xml-stylesheet type="text/css" href="gpml.css" ?>
line 3: <!DOCTYPE set SYSTEM "gpml.merged.dtd">
line 5: <import resource="GENIAontology.daml" prefix="G"></import>
java -cp <classpath> data.pos.training.GeniaPosTrainingDataExtractor GENIAcorpus4 GENIAcorpus4.02.pos.xmlxml  <genia-pos-training-data>

...

Creating a model

java -cp <classpath>  opennlp.tools.postag.POSTaggerME  <training-data>  <model-name>  iterations  cutoff
Where

• <training-data>* is an OpenNLP training data file.
• <model-name>* is the file name of the resulting model. The name should end with either .txt (for a plain text model) or .bin.gz (for a compressed binary model).
• <iterations>* determines how many training iterations will be performed. The default is 100.
• <cutoff>* determines the minimum number of times a feature has to be seen to be considered for inclusion in the model. The default cutoff is 5

...

We have provided a mechanism for creating a tag dictionary. It can be run with the following command:
java -cp <classpath>edu<classpath>  org.mayoapache.bmictakes.uima.pos_tagger.TagDictionaryCreatorpostagger.TagDictionaryCreator  <training-data>  <tag-dictionary>dictionary>  case-sensitive
Where

• <training-data>* is a file containing pos-of-speech tagged training data
• <tag-dictionary>* the file name of the resulting tag dictionary
• <case-sensitive>* is either 'true' or 'false' depending on whether the tag dictionary should be case sensitive or not.

...

OpenNLP provides a default tag dictionary for the English part-of-speech model called tag.bin.gz which can be downloaded from
}+from http://opennlp.sourceforge.net/models/english/parser/tagdict+.
. You  You should use this tag dictionary only if you are using the model from
+ http://opennlp.sourceforge.net/models/english/parser/tag.bin.gz+.

...

If this is gold standard sentence:
  The_DT major_JJ inducible_JJ protein_NN complex_NN that_WDT binds_VBZ ._.

And if this is the output for that sentence:
  The_DT major_JJ inducible_NN protein_NN complex_NN that_WDT binds_VBD ._.

...

• Use tokenizer generated tokens
• Run the tokenizer and use this as input to the POS tagger.
• In this scenario, we calculate F-measure in the following way:

true positive (TP)
  a token that has the correct boundary and part-of-speech label

false positive (FP)
  a tagged token that does not have the correct boundary and/or part-of-speech label

false negative (FN)
  a token in the gold standard data that was not correctly generated by the tokenizer/POS tagger

An example is given in
Background Color

...

given below in "Evaluate a POS tagger using generated tokens"

Evaluate a POS tagger using generated tokens

...

TP = 4, FP = 2, and FN = 3
F-measure = (2 * recall * precision) / (precision + recall) = (2 * TP) /
 (2*TP + FP + FN) = (2 * 4) / (2*4 + 2 + 3) = 8 / 13 = .615

In fact, if you do the evaluation this way for the "gold standard tokens" evaluation, then you will get the same answer as the accuracy calculation given above.