Package cc.mallet.extract

Class Extraction

java.lang.Object

cc.mallet.extract.Extraction

public class Extraction
extends java.lang.Object

The results of doing information extraction. This is designed to handle field extraction from a single document, or relation extraction and coreference from multiple documents;

Constructor Summary

Constructors

Constructor	Description
Extraction(Extractor extractor, LabelAlphabet dict)	Creates an empty Extraction option.
Extraction(Extractor extractor, LabelAlphabet dict, java.lang.String name, Tokenization input, Sequence output, java.lang.String background)	Creates an extration given a sequence output by some kind of per-sequece labeler, like an HMM or a CRF.

Method Summary

Modifier and Type	Method	Description
void	addDocumentExtraction(DocumentExtraction docseq)
void	cleanFields(FieldCleaner cleaner)
DocumentExtraction	getDocumentExtraction(int idx)
Extractor	getExtractor()
LabelAlphabet	getLabelAlphabet()
int	getNumDocuments()
int	getNumRecords()
Record	getRecord(int idx)
Record	getTargetRecord(int docnum)
void	print(java.io.PrintWriter writer)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Extraction

public Extraction(Extractor extractor,
                  LabelAlphabet dict)

Creates an empty Extraction option. DocumentExtractions can be added later by the addDocumentExtraction method.

Extraction

public Extraction(Extractor extractor,
                  LabelAlphabet dict,
                  java.lang.String name,
                  Tokenization input,
                  Sequence output,
                  java.lang.String background)

Creates an extration given a sequence output by some kind of per-sequece labeler, like an HMM or a CRF. The extraction will contain a single document.

Method Detail

addDocumentExtraction

public void addDocumentExtraction(DocumentExtraction docseq)

getRecord

public Record getRecord(int idx)

getNumRecords

public int getNumRecords()

getDocumentExtraction

public DocumentExtraction getDocumentExtraction(int idx)

getNumDocuments

public int getNumDocuments()

getExtractor

public Extractor getExtractor()

getTargetRecord

public Record getTargetRecord(int docnum)

getLabelAlphabet

public LabelAlphabet getLabelAlphabet()

cleanFields

public void cleanFields(FieldCleaner cleaner)

print

public void print(java.io.PrintWriter writer)