eus.ixa.ixa.pipe.nerc

Class StatisticalNameFinder

java.lang.Object

eus.ixa.ixa.pipe.nerc.StatisticalNameFinder

All Implemented Interfaces:

NameFinder

public class StatisticalNameFinder
extends Object
implements NameFinder

Statistical Sequence Labeling based on Apache OpenNLP Machine Learning API.

Version:

2015-09-17

Author:

ragerri

Constructor Summary

Constructors

Constructor and Description
StatisticalNameFinder(Properties props) Construct a probabilistic name finder specifying lang, model and beamsize.
StatisticalNameFinder(Properties props, NameFactory aNameFactory) Construct a StatisticalNameFinder specifying the language, a name factory, the model, the features and the beam size for decoding.

Method Summary

Modifier and Type	Method and Description
void	clearAdaptiveData() Forgets all adaptive data which was collected during previous calls to one of the find methods.
List<Name>	getNames(String[] tokens) Method to produce a list of the Name objects classified by the probabilistic model.
List<Name>	getNamesFromSpans(opennlp.tools.util.Span[] neSpans, String[] tokens) Creates a list of Name objects from spans and tokens.
opennlp.tools.util.Span[]	nercToSpans(String[] tokens) This method receives as input an array of tokenized text and calls the NameFinderME.find(tokens) to recognize and classify Named Entities.

Methods inherited from class java.lang.Object

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

Constructor Detail

StatisticalNameFinder

public StatisticalNameFinder(Properties props)

Construct a probabilistic name finder specifying lang, model and beamsize.

Parameters:

props - the properties to be loaded

StatisticalNameFinder

public StatisticalNameFinder(Properties props,
                             NameFactory aNameFactory)

Construct a StatisticalNameFinder specifying the language, a name factory, the model, the features and the beam size for decoding.

Parameters:

props - the properties

aNameFactory - the name factory to construct Name objects

Method Detail

getNames

public final List<Name> getNames(String[] tokens)

Method to produce a list of the Name objects classified by the probabilistic model. Takes an array of tokens, calls nercToSpans function for probabilistic NERC and returns a List of Name objects containing the nameString, the type and the Span

Specified by:

getNames in interface NameFinder

Parameters:

tokens - an array of tokenized text

Returns:

a List of names

nercToSpans

public final opennlp.tools.util.Span[] nercToSpans(String[] tokens)

This method receives as input an array of tokenized text and calls the NameFinderME.find(tokens) to recognize and classify Named Entities. It outputs the spans of the detected and classified Named Entities. From Apache OpenNLP documentation: "After every document clearAdaptiveData must be called to clear the adaptive data in the feature generators. Not calling clearAdaptiveData can lead to a sharp drop in the detection rate after a few documents."

Specified by:

nercToSpans in interface NameFinder

Parameters:

tokens - an array of tokenized text

Returns:

an list of Spans of Named Entities

getNamesFromSpans

public final List<Name> getNamesFromSpans(opennlp.tools.util.Span[] neSpans,
                                          String[] tokens)

Creates a list of Name objects from spans and tokens.

Specified by:

getNamesFromSpans in interface NameFinder

Parameters:

neSpans - the named entity spans of a sentence

tokens - the tokens in the sentence

Returns:

a list of Name objects

clearAdaptiveData

public final void clearAdaptiveData()

Forgets all adaptive data which was collected during previous calls to one of the find methods. This method is typically called at the end of a document. From Apache OpenNLP documentation: "After every document clearAdaptiveData must be called to clear the adaptive data in the feature generators. Not calling clearAdaptiveData can lead to a sharp drop in the detection rate after a few documents."

Specified by:

clearAdaptiveData in interface NameFinder