Contents:
- Included sentence detector model 
- Training a sentence detector model from the included sample data  
- Training a sentence detector model from other training data  
- Verifying you can train a sentence detector model successfully          
- FYI - Using opennlp directly
- Running the SentenceDetectorAnnotator analysis engine
	- SentenceDetectorAggregate.xml
	- SentenceDetectorCPE.xml
	- SentenceDetectorAnnotator.xml


################################################################
Included sentence detector model  
################################################################
A sentence detector model is included with this project.

The model derives from a combination of GENIA, Penn Treebank (Wall Street Journal) and anonymized
clinical data per Safe Harbor HIPAA guidelines. Prior to model building, the clinical data was 
deidentified for patient names to preserve patient confidentiality. Any person name in the model 
will originate from non-patient data sources.


################################################################
Training a sentence detector model from the included sample data  
################################################################
   
To train a sentence detector that recognizes the same set of candidate
end-of-sentence characters that the SentenceDetectorAnnotator uses, 
using the included sample training sentences, use the included launch 
called "SentenceDetector - train a new model", or use:
  java edu.mayo.bmi.uima.core.ae.SentenceDetector 
        data/test/sample_sd_training_sentences.txt 
        resources/sentdetect/sample_sd.mod
        

###########################################################
Training a sentence detector model from other training data 
###########################################################

To train a sentence detector that recognizes the same set of candidate
end-of-sentence characters that the SentenceDetectorAnnotator uses, 
on the Run dialog within eclipse, set the Program arguments on the Arguments 
tab (the first argument is your sentences training data file) for the 
included launch called "SentenceDetector - train a new model".

The first argument is your sentences training data file
The second argument is the name of the model file to be created
The third argument is the number of iterations for training
The fourth argument is the cutoff value used

An alternative to using the launch is to use:
  java edu.mayo.bmi.uima.core.ae.SentenceDetector sents_file model iters? cut?
where sents_file contains one sentence per line,
model is the file that is to be created, 
and iters and cut are optional parameters.

For example, sents_file could contain the following four lines:
The boy ran.
Did the girl run too?
Yes, she did.
Where did she go?


##############################################################
Verifying you can train a sentence detector model successfully          
##############################################################

Train a sentence detector model from the included sample data, as described 
above.
Compare the contents of your sentence detector model to the included sample
sentence detector model.  You might use diff, like this:
diff  resources/sentdetect/sample_sd.mod   resources/sentdetect/sample_sd_included.mod

        
############################
FYI - Using opennlp directly
############################
 
You can train a sentence detector directly using the opennlp sentence detector 
(SentenceDetectorME) with the default set of candidate end-of-sentence 
characters, using 
  java opennlp.tools.sentdetect.SentenceDetectorME infile outfile (iters cut)?
where infile contains one sentence per line (new lines separating sentences)  
For example, infile could contain the following four lines:
The boy ran.
Did the girl run too?
Yes, she did.
Where did she go?


######################################################
Running the SentenceDetectorAnnotator analysis engine
######################################################

Launch the "SentenceDetector annotator" run configuration.
Open CPE SentenceDetectorCPE.xml, which uses the analysis
engines listed in SentenceDetectorAggregate.xml.

The aggregate analysis engine is defined to read from plain text file(s) in 
data\test\sample_notes_plaintext
using the FilesInDirectoryCollectionReader.

Since the sentence annotator processes the text one section at a time, there 
must be at least one section (segment) annotation for the SentenceDetectorAnnotator
to add Sentence annotations. 
Therefore the first analysis engine is the SimpleSegmentAnnotator, which
creates a single Segment annotation that covers the entire text.
Then the SentenceDetectorAnnotator analysis engine adds Sentence annotations.