eus.ixa.ixa.pipe.nerc.dict

Class MFSResource

java.lang.Object

eus.ixa.ixa.pipe.nerc.dict.MFSResource

All Implemented Interfaces:

opennlp.tools.util.model.SerializableArtifact

public class MFSResource
extends Object
implements opennlp.tools.util.model.SerializableArtifact

Reads wordnet lexicons formated as house#n\t1092#noun.artifact to search for most frequent senses.

Version:

2015-03-30

Author:

ragerri

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	MFSResource.MFSResourceSerializer

Constructor Summary

Constructors

Constructor and Description
MFSResource(InputStream in) Build the MFS Dictionary.

Method Summary

Modifier and Type	Method and Description
Class<?>	getArtifactSerializerClass()
List<String>	getFirstSenseBilou(List<String> lemmas, String[] posTags) Extract most frequent sense baseline from WordNet data, using Ciaramita and Altun's (2006) approach for bilou encoding.
List<String>	getFirstSenseBio(List<String> lemmas, String[] posTags) Extract most frequent sense baseline from WordNet data, using Ciaramita and Altun's (2006) approach for a bio encoding.
String	getMFS(com.google.common.collect.TreeMultimap<Integer,String> mfsMap) Get the MFS from a lemma#posClass entry, e.g., house#n.
com.google.common.collect.TreeMultimap<Integer,String>	getMFSRanking(String lemmaPOSClass, Integer rankSize) Get a rank of senses ordered by MFS.
com.google.common.collect.TreeMultimap<Integer,String>	getOrderedMap(String lemmaPOSClass) Get the ordered Map of most frequent senses for a lemma#pos entry.
void	getOrderedSenses(List<String> mfsList, com.google.common.collect.TreeMultimap<Integer,String> mfsResultsMap) Look-up lemma#pos string as key in dictionary.
void	serialize(OutputStream out) Serialize the lexicon in the original format.

Methods inherited from class java.lang.Object

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

Constructor Detail

MFSResource

public MFSResource(InputStream in)
            throws IOException

Build the MFS Dictionary.

Parameters:

in - the input stream

Throws:

IOException - the io exception

Method Detail

getFirstSenseBio

public List<String> getFirstSenseBio(List<String> lemmas,
                                     String[] posTags)

Extract most frequent sense baseline from WordNet data, using Ciaramita and Altun's (2006) approach for a bio encoding.

Parameters:

lemmas - in the sentence

posTags - the postags of the sentence

Returns:

the most frequent senses for the sentence

getFirstSenseBilou

public List<String> getFirstSenseBilou(List<String> lemmas,
                                       String[] posTags)

Extract most frequent sense baseline from WordNet data, using Ciaramita and Altun's (2006) approach for bilou encoding.

Parameters:

lemmas - in the sentence

posTags - posTags in the sentence

Returns:

the most frequent senses for the sentence

getOrderedMap

public com.google.common.collect.TreeMultimap<Integer,String> getOrderedMap(String lemmaPOSClass)

Get the ordered Map of most frequent senses for a lemma#pos entry.

Parameters:

lemmaPOSClass - the lemma#pos entry

Returns:

the ordered multimap of senses

getOrderedSenses

public void getOrderedSenses(List<String> mfsList,
                             com.google.common.collect.TreeMultimap<Integer,String> mfsResultsMap)

Look-up lemma#pos string as key in dictionary.

Parameters:

mfsList - the list containing the freq#sense values

mfsResultsMap - the map in which the results are stored

getMFS

public String getMFS(com.google.common.collect.TreeMultimap<Integer,String> mfsMap)

Get the MFS from a lemma#posClass entry, e.g., house#n.

Parameters:

mfsMap - map to get the MFS from

Returns:

the most frequent sense

getMFSRanking

public com.google.common.collect.TreeMultimap<Integer,String> getMFSRanking(String lemmaPOSClass,
                                                                            Integer rankSize)

Get a rank of senses ordered by MFS.

Parameters:

lemmaPOSClass - the lemma#pos entry

rankSize - the size of the rank

Returns:

the ordered multimap containing the rank

serialize

public void serialize(OutputStream out)
               throws IOException

Serialize the lexicon in the original format.

Parameters:

out - the output stream

Throws:

IOException - if io errors

getArtifactSerializerClass

public Class<?> getArtifactSerializerClass()

Specified by:

getArtifactSerializerClass in interface opennlp.tools.util.model.SerializableArtifact