Revision: 17101
http://sourceforge.net/p/gate/code/17101
Author: valyt
Date: 2013-11-18 19:30:57 +0000 (Mon, 18 Nov 2013)
Log Message:
-----------
Started work on a text classification plugin, using Semantic Spaces to produce
context vectors, and a LIBLINEAR model for the actual classification.
Added Paths:
-----------
gate/trunk/plugins/Text_Classification/
gate/trunk/plugins/Text_Classification/.classpath
gate/trunk/plugins/Text_Classification/.project
gate/trunk/plugins/Text_Classification/build/
gate/trunk/plugins/Text_Classification/build/ivy.xml
gate/trunk/plugins/Text_Classification/build/ivysettings.xml
gate/trunk/plugins/Text_Classification/build.xml
gate/trunk/plugins/Text_Classification/creole.xml
gate/trunk/plugins/Text_Classification/src/
gate/trunk/plugins/Text_Classification/src/gate/
gate/trunk/plugins/Text_Classification/src/gate/ml/
gate/trunk/plugins/Text_Classification/src/gate/ml/textclass/
gate/trunk/plugins/Text_Classification/src/gate/ml/textclass/TextClassificationPR.java
Index: gate/trunk/plugins/Text_Classification
===================================================================
--- gate/trunk/plugins/Text_Classification 2013-11-18 02:22:49 UTC (rev
17100)
+++ gate/trunk/plugins/Text_Classification 2013-11-18 19:30:57 UTC (rev
17101)
Property changes on: gate/trunk/plugins/Text_Classification
___________________________________________________________________
Added: svn:ignore
## -0,0 +1,3 ##
+classes
+textClassification.jar
+.settings
Added: gate/trunk/plugins/Text_Classification/.classpath
===================================================================
--- gate/trunk/plugins/Text_Classification/.classpath
(rev 0)
+++ gate/trunk/plugins/Text_Classification/.classpath 2013-11-18 19:30:57 UTC
(rev 17101)
@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<classpath>
+ <classpathentry kind="src" path="src"/>
+ <classpathentry kind="con"
path="org.apache.ivyde.eclipse.cpcontainer.IVYDE_CONTAINER/?project=gate-plugin-text-classification&ivyXmlPath=build%2Fivy.xml&confs=*"/>
+ <classpathentry combineaccessrules="false" kind="src" path="/GATE"/>
+ <classpathentry kind="con"
path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.6"/>
+ <classpathentry kind="output" path="classes"/>
+</classpath>
Property changes on: gate/trunk/plugins/Text_Classification/.classpath
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Added: gate/trunk/plugins/Text_Classification/.project
===================================================================
--- gate/trunk/plugins/Text_Classification/.project
(rev 0)
+++ gate/trunk/plugins/Text_Classification/.project 2013-11-18 19:30:57 UTC
(rev 17101)
@@ -0,0 +1,18 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+ <name>gate-plugin-text-classification</name>
+ <comment></comment>
+ <projects>
+ </projects>
+ <buildSpec>
+ <buildCommand>
+ <name>org.eclipse.jdt.core.javabuilder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ </buildSpec>
+ <natures>
+ <nature>org.eclipse.jdt.core.javanature</nature>
+ <nature>org.apache.ivyde.eclipse.ivynature</nature>
+ </natures>
+</projectDescription>
Property changes on: gate/trunk/plugins/Text_Classification/.project
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Index: gate/trunk/plugins/Text_Classification/build
===================================================================
--- gate/trunk/plugins/Text_Classification/build 2013-11-18 02:22:49 UTC
(rev 17100)
+++ gate/trunk/plugins/Text_Classification/build 2013-11-18 19:30:57 UTC
(rev 17101)
Property changes on: gate/trunk/plugins/Text_Classification/build
___________________________________________________________________
Added: svn:ignore
## -0,0 +1 ##
+ivy-report
Added: gate/trunk/plugins/Text_Classification/build/ivy.xml
===================================================================
--- gate/trunk/plugins/Text_Classification/build/ivy.xml
(rev 0)
+++ gate/trunk/plugins/Text_Classification/build/ivy.xml 2013-11-18
19:30:57 UTC (rev 17101)
@@ -0,0 +1,18 @@
+<ivy-module version="2.0">
+
+ <info
+ organisation="uk.ac.gate"
+ module="gate-plugin-text-classification"
+ revision="0.1-SNAPSHOT">
+ <description homepage="http://gate.ac.uk/"; />
+ </info>
+
+ <configurations>
+ <conf name="default" />
+ </configurations>
+
+ <dependencies defaultconf="*->master(default),runtime(default)" >
+ <dependency org="edu.ucla.sspace" name="sspace" rev="2.0.4"/>
+ <dependency org="de.bwaldvogel" name="liblinear" rev="1.92"/>
+ </dependencies>
+</ivy-module>
Added: gate/trunk/plugins/Text_Classification/build/ivysettings.xml
===================================================================
--- gate/trunk/plugins/Text_Classification/build/ivysettings.xml
(rev 0)
+++ gate/trunk/plugins/Text_Classification/build/ivysettings.xml
2013-11-18 19:30:57 UTC (rev 17101)
@@ -0,0 +1,10 @@
+<ivysettings>
+ <settings defaultResolver="default"/>
+ <resolvers>
+ <chain name="default" returnFirst="true">
+ <ibiblio name="maven-central" m2compatible="true" />
+ <ibiblio name="gate.ac.uk" m2compatible="true"
+ root="http://repo.gate.ac.uk/content/groups/public/"; />
+ </chain>
+ </resolvers>
+</ivysettings>
\ No newline at end of file
Added: gate/trunk/plugins/Text_Classification/build.xml
===================================================================
--- gate/trunk/plugins/Text_Classification/build.xml
(rev 0)
+++ gate/trunk/plugins/Text_Classification/build.xml 2013-11-18 19:30:57 UTC
(rev 17101)
@@ -0,0 +1,150 @@
+<project name="gate-plugin-text-classification" default="dist" basedir="."
+ xmlns:ivy="antlib:org.apache.ivy.ant">
+ <!-- Prevent Ant from warning about includeantruntime not being set -->
+ <property name="build.sysclasspath" value="ignore" />
+
+ <property file="build.properties" />
+
+ <property name="src" location="src"/>
+ <property name="build" location="classes"/>
+ <property name="dist" location="dist"/>
+ <property name="jar.location" location="textClassification.jar" />
+ <property name="test.jar.location" location="mltests.jar" />
+
+ <property name="gate.home" location="../.." />
+ <property name="gate.lib" location="${gate.home}/lib" />
+ <property name="gate.jar" location="${gate.home}/bin/gate.jar" />
+ <property name="docDir" location="doc" />
+
+ <!-- Load Ivy support -->
+ <property name="ivy.dir" location="build" />
+ <path id="ivy.lib.path">
+ <fileset file="${gate.home}/build/lib/ivy.jar" />
+ </path>
+ <taskdef resource="org/apache/ivy/ant/antlib.xml"
+ uri="antlib:org.apache.ivy.ant"
+ classpathref="ivy.lib.path"/>
+
+
+ <!-- Just the core GATE classes -->
+ <path id="gate.classpath">
+ <pathelement location="${gate.jar}" />
+ <fileset dir="${gate.lib}">
+ <include name="**/*.jar" />
+ <include name="**/*.zip" />
+ </fileset>
+ </path>
+
+ <!-- GATE classes plus local lib jars -->
+ <path id="compile.classpath">
+ <path refid="gate.classpath" />
+ <path refid="ivylibs" />
+ </path>
+
+ <target name="init">
+ <!-- Create the time stamp -->
+ <tstamp/>
+ <!-- Create the build directory structure used by compile -->
+ <mkdir dir="${build}"/>
+ </target>
+
+ <target name="depend">
+ <ivy:configure file="${ivy.dir}/ivysettings.xml"/>
+ <ivy:resolve file="${ivy.dir}/ivy.xml" log="quiet"/>
+ <ivy:report todir="${ivy.dir}/ivy-report" />
+ <ivy:cachepath pathid="ivylibs"/>
+ </target>
+
+ <target name="compile" depends="init, depend"
+ description="compile the source " >
+ <!-- Compile the java code from ${src} into ${build} -->
+ <javac srcdir="${src}" destdir="${build}" debug="true" source="1.6"
target="1.6">
+ <classpath refid="compile.classpath"/>
+ </javac>
+ </target>
+
+ <target name="copy.resources" depends="init"
+ description="copy non-.java files from src to build" >
+ <copy todir="${build}" includeEmptyDirs="false">
+ <fileset dir="${src}" excludes="**/*.java" />
+ </copy>
+ </target>
+
+ <!-- build a JAR containing everything except the test classes -->
+ <target name="dist" depends="compile, copy.resources"
+ description="generate the distribution" >
+ <jar jarfile="${jar.location}" >
+ <fileset dir="${build}" excludes="**/Test*" />
+ </jar>
+ </target>
+
+ <target name="clean.classes"
+ description="clean up" >
+ <!-- Delete the ${build} and ${dist} directory trees -->
+ <delete dir="${build}"/>
+ <delete dir="${dist}"/>
+ </target>
+
+ <target name="clean" depends="clean.classes" >
+ <!-- full clean also removes the generated JAR file -->
+ <delete file="${jar.location}" />
+ <delete file="${test.jar.location}" />
+ </target>
+
+ <!-- build a jar containing just the test classes -->
+ <target name="testjar" depends="compile, copy.resources">
+ <jar jarfile="${test.jar.location}">
+ <fileset dir="${build}" includes="**/Test*" />
+ </jar>
+ </target>
+
+ <target name="test" depends="dist, testjar">
+ <!-- note that compiled JAR file and local libs are deliberately NOT
+ included on classpath - they are loaded through the GATE
+ classloader -->
+ <junit fork="true" showoutput="yes" printsummary="yes">
+ <classpath>
+ <pathelement location="${test.jar.location}" />
+ <path refid="gate.classpath" />
+ </classpath>
+ <sysproperty key="gate.home.location" value="${gate.home}" />
+ <sysproperty key="machinelearning.plugin" file="." />
+ <sysproperty key="annie.plugin" file="../ANNIE" />
+ <!-- don't load any plugins automatically -->
+ <sysproperty key="load.plugin.path" value=";" />
+ <!-- run headless -->
+ <sysproperty key="java.awt.headless" value="true" />
+ <formatter type="xml" />
+ <test name="gate.creole.ml.maxent.TestMaxentWrapper" />
+ </junit>
+ </target>
+
+ <target name="javadoc">
+ <javadoc access="protected"
+ destdir="${docDir}/javadoc"
+ classpathref="compile.classpath"
+ Encoding="UTF-8"
+ Use="yes"
+ Windowtitle="GATE learning plugin JavaDoc"
+ docencoding="UTF-8"
+ charset="UTF-8"
+ source="1.6"
+ useexternalfile="yes"
+ breakiterator="true">
+ <fileset dir="${src}"/>
+ <link href="http://docs.oracle.com/javase/6/docs/api/"; />
+ <link href="http://gate.ac.uk/gate/doc/javadoc/"; />
+ </javadoc>
+ </target>
+
+ <!-- Targets used by the main GATE build file -->
+ <target name="build" depends="dist" />
+
+ <target name="distro.prepare" depends="clean.classes">
+ <!-- also remove JUnit test reports -->
+ <delete>
+ <fileset dir="." includes="TEST*.xml" />
+ </delete>
+ <delete file="${test.jar.location}" />
+ </target>
+</project>
Property changes on: gate/trunk/plugins/Text_Classification/build.xml
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Added: gate/trunk/plugins/Text_Classification/creole.xml
===================================================================
--- gate/trunk/plugins/Text_Classification/creole.xml
(rev 0)
+++ gate/trunk/plugins/Text_Classification/creole.xml 2013-11-18 19:30:57 UTC
(rev 17101)
@@ -0,0 +1,5 @@
+<?xml version="1.0"?>
+<!-- $Id: creole.xml 11850 2009-10-30 15:22:04Z ian_roberts $ -->
+<CREOLE-DIRECTORY>
+ <JAR SCAN="true">textClassification.jar</JAR>
+</CREOLE-DIRECTORY>
Added:
gate/trunk/plugins/Text_Classification/src/gate/ml/textclass/TextClassificationPR.java
===================================================================
---
gate/trunk/plugins/Text_Classification/src/gate/ml/textclass/TextClassificationPR.java
(rev 0)
+++
gate/trunk/plugins/Text_Classification/src/gate/ml/textclass/TextClassificationPR.java
2013-11-18 19:30:57 UTC (rev 17101)
@@ -0,0 +1,451 @@
+/**
+ *
+ */
+package gate.ml.textclass;
+
+import java.io.BufferedInputStream;
+import java.io.BufferedReader;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.StringReader;
+import java.io.UnsupportedEncodingException;
+import java.net.URISyntaxException;
+import java.net.URL;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Properties;
+import java.util.Set;
+
+import de.bwaldvogel.liblinear.Feature;
+import de.bwaldvogel.liblinear.FeatureNode;
+import de.bwaldvogel.liblinear.Linear;
+import de.bwaldvogel.liblinear.Model;
+import edu.ucla.sspace.common.DocumentVectorBuilder;
+import edu.ucla.sspace.common.SemanticSpace;
+import edu.ucla.sspace.common.SemanticSpaceIO;
+import edu.ucla.sspace.vector.DenseVector;
+import edu.ucla.sspace.vector.DoubleVector;
+import gate.Annotation;
+import gate.AnnotationSet;
+import gate.LanguageAnalyser;
+import gate.Resource;
+import gate.Utils;
+import gate.creole.AbstractLanguageAnalyser;
+import gate.creole.ExecutionException;
+import gate.creole.ResourceInstantiationException;
+import gate.creole.metadata.CreoleParameter;
+import gate.creole.metadata.CreoleResource;
+import gate.creole.metadata.RunTime;
+
+/**
+ * A simple text classification PR, using a <a
+ * href="https://github.com/fozziethebeat/S-Space/";>Semantic Space</a>
+ * implementation to generate vectors for the annotations to be classified,
+ * and a <a href="http://www.csie.ntu.edu.tw/~cjlin/liblinear";>LIBLINEAR</a>
+ * model to perform the actual classification.
+ */
+@CreoleResource
+public class TextClassificationPR extends AbstractLanguageAnalyser implements
+
LanguageAnalyser {
+
+ /**
+ * Serialisation UID.
+ */
+ private static final long serialVersionUID = 141243024466478134L;
+
+ /**
+ * The semantic space used to generate 'document' feature vectors for the
+ * annotations to be classified.
+ */
+ protected SemanticSpace semanticSpace;
+
+ protected DocumentVectorBuilder vectorBuilder;
+
+ /**
+ * The LibLinear model used for classification.
+ */
+ protected Model libLinearModel;
+
+ /**
+ * Thresholds to be used when classifying text annotations. If the model
+ * classifies an input annotation as class "X", and "X"
+ * exists as a key in this map, then the classification is only effected if
+ * the classification probability emitted by the model is greater than or
+ * equal to the value in the map. Values in this map are expected to be
+ * probabilities, i.e. positive values between 0.0 and 1.0.
+ *
+ * Note that the liblinear model used must be able to produce classification
+ * probabilities. At the time of writing, only Linear Regression models are
+ * able to do so, while SVM-based one are not. Models that cannot supply
+ * probabilities will simply return 1.0 as the classification probability,
+ * causing values in this map to have no effect.
+ */
+ protected Map<String, Double> customClassThresholds;
+
+ /**
+ * The set of stop-words to be used. The values in this list are matched
+ * verbatim against the values of the configured input feature on the input
+ * annotations.
+ */
+ protected Set<String> stopWords;
+
+ /**
+ * URL to a file containing stop words, one on each line.
+ * Lines starting with # or // are considered comments and are ignored.
+ * White space at the start and end of each line's content will be ignored.
+ */
+ protected URL stopWordsURL;
+
+ /**
+ * The name for the annotation set used for input.
+ */
+ protected String inputASName;
+
+ /**
+ * The type of input annotations.
+ */
+ protected String inputAnnotationType;
+
+ /**
+ * The type for token annotations.
+ */
+ protected String tokenAnnotationType;
+
+ /**
+ * The feature on token annotations used to collect the input text.
+ */
+ protected String inputFeatureName;
+
+ /**
+ * The name for the output annotation set.
+ */
+ protected String outputASName;
+
+ /**
+ * The type of annotations produced. If this is the same as
+ * inputAnnotationType, and the input and output annotation sets are the
same,
+ * then no new annotations will be created. The output feature value will
+ * simply be added to the input annotations.
+ */
+ protected String outputAnnotationType;
+
+ /**
+ * The name feature of the feature on the output annotations that stores the
+ * class.
+ */
+ protected String outputFeatureName;
+
+ /**
+ * URL for the file containing the serialized semantic space.
+ */
+ protected URL sematicSpaceURL;
+
+ /**
+ * The URL to the LIBLINEAR model used for classification.
+ */
+ protected URL modelURL;
+
+
+ @Override
+ public Resource init() throws ResourceInstantiationException {
+ // load the model
+ if(modelURL != null) {
+ InputStreamReader isr = null;
+ try{
+ // the default implementation uses ISO-8859-1, so we do the same
+ isr = new InputStreamReader(
+ new BufferedInputStream(modelURL.openStream()), "ISO-8859-1");
+ libLinearModel = Linear.loadModel(isr);
+ isr.close();
+ } catch(IOException ioe) {
+ throw new ResourceInstantiationException(
+ "IO Error while loading the model from " + modelURL, ioe);
+ }
+ } else throw new ResourceInstantiationException("No model URL provided.");
+ // load the semantic space
+ if(sematicSpaceURL != null) {
+ if(sematicSpaceURL.getProtocol().equalsIgnoreCase("file")) {
+ try {
+ semanticSpace = SemanticSpaceIO.load(sematicSpaceURL.getFile());
+ } catch(IOException e) {
+ throw new ResourceInstantiationException(
+ "I/O error while loading the semantic space from " +
+ sematicSpaceURL, e);
+ }
+ } else {
+ throw new ResourceInstantiationException(
+ "URL provided for the semantic space file (" + sematicSpaceURL +
+ ")was not to a local file.");
+ }
+ } else {
+ throw new ResourceInstantiationException(
+ "No URL provided for the semantic space file.");
+ }
+
+ // create the vector builder
+ Properties config = new Properties();
+ config.put(DocumentVectorBuilder.USE_TERM_FREQUENCIES_PROPERTY, true);
+ vectorBuilder = new DocumentVectorBuilder(semanticSpace, config);
+
+ // load the stop words
+ stopWords = null;
+ if(stopWordsURL != null) {
+ try {
+ BufferedReader swReader = new BufferedReader(
+ new InputStreamReader(stopWordsURL.openStream(), "UTF-8"));
+ stopWords = new HashSet<String>();
+ String line = swReader.readLine();
+ while(line != null) {
+ line = line.trim();
+ if(line.startsWith("#") || line.startsWith("//")) {
+ //ignore comment
+ } else {
+ stopWords.add(line);
+ }
+ line = swReader.readLine();
+ }
+ } catch(IOException e) {
+ throw new ResourceInstantiationException(
+ "I/O error while reading the stop words.", e);
+ }
+ }
+ return super.init();
+ }
+
+ @Override
+ public void execute() throws ExecutionException {
+ // normalize parameters
+ if(inputASName == null || inputASName.length() == 0) inputASName = "";
+ if(outputASName == null || outputASName.length() == 0) outputASName = "";
+ // should we use the input annotations for output
+ boolean sameAnnotation = inputASName.equals(outputASName) &&
+ inputAnnotationType.equals(outputAnnotationType);
+ // validate parameter values
+ if(inputAnnotationType == null || inputAnnotationType.length() == 0) {
+ throw new ExecutionException("No input annotation type provided.");
+ }
+ if(inputFeatureName == null || inputFeatureName.length() == 0) {
+ throw new ExecutionException("No input feature name provided.");
+ }
+ if(tokenAnnotationType == null || tokenAnnotationType.length() == 0) {
+ throw new ExecutionException("No token annotation type provided.");
+ }
+ if(outputAnnotationType == null || outputAnnotationType.length() == 0) {
+ throw new ExecutionException("No output annotation type provided.");
+ }
+ if(outputFeatureName == null || outputFeatureName.length() == 0) {
+ throw new ExecutionException("No output feature name provided.");
+ }
+
+
+ // collect instance annotations
+ AnnotationSet inputAS = document.getAnnotations(inputASName);
+ AnnotationSet instances = inputAS.get(inputAnnotationType);
+ for(Annotation instAnn : instances) {
+ // collect the tokens
+ List<Annotation> instTokens = Utils.inDocumentOrder(
+ Utils.getContainedAnnotations(inputAS, instAnn,
tokenAnnotationType));
+ // create a new context
+ StringBuilder instanceStrBld = new StringBuilder();
+ boolean first = true;
+ for(Annotation token : instTokens) {
+ String tokenString = (String)token.getFeatures().get(inputFeatureName);
+ if(tokenString != null && tokenString.length() > 0) {
+ if(first) {
+ first = false;
+ } else {
+ instanceStrBld.append(' ');
+ }
+ instanceStrBld.append(tokenString);
+ }
+ }
+
+ String instanceText = instanceStrBld.toString();
+ if(instanceText.length() > 0) {
+ DoubleVector instanceVector = vectorBuilder.buildVector(
+ new BufferedReader(new StringReader(instanceText)),
+ new DenseVector(semanticSpace.getVectorLength()));
+ // classify the vector
+ Feature[] features = new Feature[libLinearModel.getNrFeature()];
+ for(int i = 0; i < features.length; i++) {
+ features[i] = new FeatureNode(i, instanceVector.get(i));
+ }
+ double[] probs = new double[libLinearModel.getNrClass()];
+ double label = Linear.predictValues(libLinearModel, features, probs);
+ // do we need to check probabilities?
+ // TODO
+ if(customClassThresholds != null && customClassThresholds.size() > 0) {
+
+ }
+ }
+
+ }
+
+ }
+
+ @Override
+ public void cleanup() {
+ // TODO Auto-generated method stub
+ super.cleanup();
+ }
+
+ public String getInputASName() {
+ return inputASName;
+ }
+
+ /**
+ * Set the name for the annotation set used for input.
+ * @param inputASName
+ */
+ @CreoleParameter(comment="The name for the annotation set used for input",
+ defaultValue = "")
+ @RunTime
+ public void setInputASName(String inputASName) {
+ this.inputASName = inputASName;
+ }
+
+ public String getInputAnnotationType() {
+ return inputAnnotationType;
+ }
+
+ /**
+ * Set the type of input annotations
+ * @param inputAnnotationType
+ */
+ @CreoleParameter(comment="The type of input annotations",
+ defaultValue="Sentence")
+ @RunTime
+ public void setInputAnnotationType(String inputAnnotationType) {
+ this.inputAnnotationType = inputAnnotationType;
+ }
+
+ public String getTokenAnnotationType() {
+ return tokenAnnotationType;
+ }
+
+ /**
+ * Set the type for token annotations.
+ * @param tokenAnnotationType
+ */
+ @CreoleParameter(comment = "The type for token annotations.",
+ defaultValue = "Token")
+ @RunTime
+ public void setTokenAnnotationType(String tokenAnnotationType) {
+ this.tokenAnnotationType = tokenAnnotationType;
+ }
+
+ public String getInputFeatureName() {
+ return inputFeatureName;
+ }
+
+ /**
+ * Set the feature on token annotations used to collect the input text.
+ * @param inputFeatureName
+ */
+ @CreoleParameter(comment =
+ "The feature on the token annotations used to collect the text for " +
+ "each input annotation.", defaultValue = "root")
+ @RunTime
+ public void setInputFeatureName(String inputFeatureName) {
+ this.inputFeatureName = inputFeatureName;
+ }
+
+ public String getOutputASName() {
+ return outputASName;
+ }
+
+ /**
+ * Set the name for the output annotation set
+ * @param outputASName
+ */
+ @CreoleParameter(comment = "The name for the output annotation set",
+ defaultValue = "")
+ @RunTime
+ public void setOutputASName(String outputASName) {
+ this.outputASName = outputASName;
+ }
+
+ public String getOutputAnnotationType() {
+ return outputAnnotationType;
+ }
+
+ /**
+ * Set the type of annotations produced. If this is the same as
+ * inputAnnotationType, and the input and output annotation sets are the
same,
+ * then no new annotations will be created. The output feature value will
+ * simply be added to the input annotations.
+ * @param outputAnnotationName
+ */
+ @CreoleParameter(defaultValue = "Sentence", comment =
+ "The type of annotations produced. If this is the same as " +
+ "inputAnnotationType, and the input and output annotation sets are
the same, " +
+ "then no new annotations will be created. The output feature value
will " +
+ "simply be added to the input annotations. ")
+ @RunTime
+ public void setOutputAnnotationType(String outputAnnotationName) {
+ this.outputAnnotationType = outputAnnotationName;
+ }
+
+ public String getOutputFeatureName() {
+ return outputFeatureName;
+ }
+
+ /**
+ * Set the name feature of the feature on the output annotations that stores
+ * the class.
+ * @param outputFeatureName
+ */
+ @CreoleParameter(defaultValue = "class", comment =
+ "The name feature of the feature on the output annotations that " +
+ "stores the class. ")
+ @RunTime
+ public void setOutputFeatureName(String outputFeatureName) {
+ this.outputFeatureName = outputFeatureName;
+ }
+
+ public URL getSematicSpaceURL() {
+ return sematicSpaceURL;
+ }
+
+ /**
+ * Set the URL for the file containing the serialized semantic space.
+ * @param sematicSpaceURL
+ */
+ @CreoleParameter(comment =
+ "URL for the file containing the serialized semantic space. This must be
a local file:// URL.")
+ public void setSematicSpaceURL(URL sematicSpaceURL) {
+ this.sematicSpaceURL = sematicSpaceURL;
+ }
+
+ public URL getModelURL() {
+ return modelURL;
+ }
+
+ /**
+ * Set the URL to the LibLinear model used for classification.
+ * @param modelURL
+ */
+ @CreoleParameter(comment = "The URL to the LIBLINEAR model used for
classification")
+ public void setModelURL(URL modelURL) {
+ this.modelURL = modelURL;
+ }
+
+ public URL getStopWordsURL() {
+ return stopWordsURL;
+ }
+
+ /**
+ * Set the URL to a file containing stop words, one on each line.
+ * @param stopWordsURL
+ */
+ @CreoleParameter(
+ comment = "URL to a file containing stop words, one on each line, using
UTF-8.")
+ public void setStopWordsURL(URL stopWordsURL) {
+ this.stopWordsURL = stopWordsURL;
+ }
+
+
+
+}
Property changes on:
gate/trunk/plugins/Text_Classification/src/gate/ml/textclass/TextClassificationPR.java
___________________________________________________________________
Added: svn:keywords
## -0,0 +1 ##
+Id
\ No newline at end of property
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
This was sent by the SourceForge.net collaborative development platform, the
world's largest Open Source development site.
------------------------------------------------------------------------------
Shape the Mobile Experience: Free Subscription
Software experts and developers: Be at the forefront of tech innovation.
Intel(R) Software Adrenaline delivers strategic insight and game-changing
conversations that shape the rapidly evolving mobile landscape. Sign up now.
http://pubads.g.doubleclick.net/gampad/clk?id=63431311&iu=/4140/ostg.clktrk
_______________________________________________
GATE-cvs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/gate-cvs
