de.julielab.topicmodeling.services

Class MalletTopicModeling

java.lang.Object

de.julielab.topicmodeling.services.MalletTopicModeling

All Implemented Interfaces:

ITopicModeling

public class MalletTopicModeling
extends Object
implements ITopicModeling

Constructor Summary

Constructors

Constructor and Description
MalletTopicModeling()

Method Summary

Modifier and Type	Method and Description
static double	computeSimilarity(double[] vectorA, double[] vectorB)
String	getId(org.apache.uima.jcas.JCas aJCas)
cc.mallet.types.TokenSequence	getLemmata(org.apache.uima.jcas.JCas aJCas)
Object[]	getVocabulary(Model model)
Map<String,List<Topic>>	inferLabel(Document doc, Model model, org.apache.commons.configuration2.XMLConfiguration xmlConfig) The argument topicWords has to be passed here due to performance reasons; this object is available in every model.malletModel by the method getTopWords()
Map<String,List<Topic>>	inferLabel(org.apache.uima.jcas.JCas cas, Model model, org.apache.commons.configuration2.XMLConfiguration xmlConfig) The argument topicWords has to be passed here due to performance reasons; this object is available in every model.malletModel by the method getTopWords()
boolean	isNotNum(String lemmaString) Filters simple numbers that does not have real semantics
boolean	isNotPunctuation(String lemmaString)
List<cc.mallet.types.TokenSequence>	jcorePreprocess(List<Document> docs)
org.apache.commons.configuration2.XMLConfiguration	loadConfig(String configFile)
cc.mallet.types.InstanceList	malletPreprocess(List<cc.mallet.types.TokenSequence> data)
void	mapMalletIdToPubmedId(List<Document> docs, Model model)
void	mapPubmedIdToMalletId(List<Document> docs, Model model)
cc.mallet.types.InstanceList	preprocess(List<Document> docs)
List<Document>	readDocuments(File file, org.apache.commons.configuration2.XMLConfiguration xmlConfig)
Model	readMalletModel(File file)
Model	readModel(String filename)
List<Document>	readXmiDb(MalletTopicModeling tm, org.apache.commons.configuration2.HierarchicalConfiguration<org.apache.commons.configuration2.tree.ImmutableNode> configuration)
void	saveMalletModel(Model model, File file)
void	saveModel(Model model, String filename)
TMSearchResult	search(Document query, Model model, org.apache.commons.configuration2.XMLConfiguration xmlConfig)
TMSearchResult	searchIndexOnly(Document query, Model model, org.apache.commons.configuration2.XMLConfiguration xmlConfig)
TMSearchResult	searchModelOnly(Document query, Model model, org.apache.commons.configuration2.XMLConfiguration xmlConfig)
Model	train(cc.mallet.types.InstanceList instances, org.apache.commons.configuration2.XMLConfiguration xmlConfig)
Model	train(List<Document> docs, org.apache.commons.configuration2.XMLConfiguration xmlConfig)

Methods inherited from class java.lang.Object

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

Constructor Detail

MalletTopicModeling

public MalletTopicModeling()

Method Detail

computeSimilarity

public static double computeSimilarity(double[] vectorA,
                                       double[] vectorB)

loadConfig

public org.apache.commons.configuration2.XMLConfiguration loadConfig(String configFile)
                                                              throws org.apache.commons.configuration2.ex.ConfigurationException

Throws:

org.apache.commons.configuration2.ex.ConfigurationException

train

public Model train(List<Document> docs,
                   org.apache.commons.configuration2.XMLConfiguration xmlConfig)

Specified by:

train in interface ITopicModeling

train

public Model train(cc.mallet.types.InstanceList instances,
                   org.apache.commons.configuration2.XMLConfiguration xmlConfig)

saveMalletModel

public void saveMalletModel(Model model,
                            File file)

saveModel

public void saveModel(Model model,
                      String filename)

Specified by:

saveModel in interface ITopicModeling

readDocuments

public List<Document> readDocuments(File file,
                                    org.apache.commons.configuration2.XMLConfiguration xmlConfig)

Specified by:

readDocuments in interface ITopicModeling

readXmiDb

public List<Document> readXmiDb(MalletTopicModeling tm,
                                org.apache.commons.configuration2.HierarchicalConfiguration<org.apache.commons.configuration2.tree.ImmutableNode> configuration)

search

public TMSearchResult search(Document query,
                             Model model,
                             org.apache.commons.configuration2.XMLConfiguration xmlConfig)

Specified by:

search in interface ITopicModeling

searchModelOnly

public TMSearchResult searchModelOnly(Document query,
                                      Model model,
                                      org.apache.commons.configuration2.XMLConfiguration xmlConfig)

searchIndexOnly

public TMSearchResult searchIndexOnly(Document query,
                                      Model model,
                                      org.apache.commons.configuration2.XMLConfiguration xmlConfig)

inferLabel

public Map<String,List<Topic>> inferLabel(Document doc,
                                          Model model,
                                          org.apache.commons.configuration2.XMLConfiguration xmlConfig)

The argument topicWords has to be passed here due to performance reasons; this object is available in every model.malletModel by the method getTopWords()

Specified by:

inferLabel in interface ITopicModeling

inferLabel

public Map<String,List<Topic>> inferLabel(org.apache.uima.jcas.JCas cas,
                                          Model model,
                                          org.apache.commons.configuration2.XMLConfiguration xmlConfig)

The argument topicWords has to be passed here due to performance reasons; this object is available in every model.malletModel by the method getTopWords()

readMalletModel

public Model readMalletModel(File file)

readModel

public Model readModel(String filename)

Specified by:

readModel in interface ITopicModeling

preprocess

public cc.mallet.types.InstanceList preprocess(List<Document> docs)

jcorePreprocess

public List<cc.mallet.types.TokenSequence> jcorePreprocess(List<Document> docs)

malletPreprocess

public cc.mallet.types.InstanceList malletPreprocess(List<cc.mallet.types.TokenSequence> data)

getLemmata

public cc.mallet.types.TokenSequence getLemmata(org.apache.uima.jcas.JCas aJCas)

getId

public String getId(org.apache.uima.jcas.JCas aJCas)

isNotNum

public boolean isNotNum(String lemmaString)

Filters simple numbers that does not have real semantics

isNotPunctuation

public boolean isNotPunctuation(String lemmaString)

mapPubmedIdToMalletId

public void mapPubmedIdToMalletId(List<Document> docs,
                                  Model model)

mapMalletIdToPubmedId

public void mapMalletIdToPubmedId(List<Document> docs,
                                  Model model)

getVocabulary

public Object[] getVocabulary(Model model)