/**
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.apache.ctakes.sideeffect.ae;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Iterator;
import java.util.Map;
import java.util.Set;
import libsvm.svm;
import libsvm.svm_model;
import libsvm.svm_node;
import org.apache.uima.analysis_engine.ResultSpecification;
import org.apache.uima.analysis_engine.annotator.AnnotatorConfigurationException;
import org.apache.uima.analysis_engine.annotator.AnnotatorContext;
import org.apache.uima.analysis_engine.annotator.AnnotatorInitializationException;
import org.apache.uima.analysis_engine.annotator.AnnotatorProcessException;
import org.apache.uima.analysis_engine.annotator.JTextAnnotator_ImplBase;
import org.apache.uima.jcas.JFSIndexRepository;
import org.apache.uima.jcas.JCas;
import org.apache.ctakes.sideeffect.type.PSESentenceFeature;
import org.apache.ctakes.sideeffect.type.SESentence;
import org.apache.ctakes.sideeffect.type.SideEffectAnnotation;
/**
* Classify side effect sentences and add them to SESentence
* @author Mayo Clinic
*
*/
public class SESentenceClassifierAnnotator extends JTextAnnotator_ImplBase {
svm_model model; //trained libsvm model
Map<String, String> feaMap; //key:nominal values, values: converted value
public void initialize(AnnotatorContext annotCtx)
throws AnnotatorInitializationException, AnnotatorConfigurationException
{
super.initialize(annotCtx);
try {
String pathOfTrainedModel = (String) getContext().getConfigParameterValue("PathOfModel");
model = svm.svm_load_model(pathOfTrainedModel);
}
catch (Exception e) {
e.printStackTrace();
}
//mapping nominal values to integer
//why convert to integer? tried binary values but didn't performed well
feaMap = new HashMap<String, String>();
feaMap.put("nul", "0"); // both side effect keyword and location features
feaMap.put("pre", "1"); // side effect keyword feature
feaMap.put("bet", "1"); // location feature
feaMap.put("bpd", "2"); // location feature
feaMap.put("bdp", "3"); // location feature
feaMap.put("bap", "4"); // location feature
feaMap.put("bep", "5"); // location feature
feaMap.put("afp", "6"); // location feature
feaMap.put("any", "7"); // location feature
}
public void process(JCas jcas, ResultSpecification resSpec)
throws AnnotatorProcessException {
JFSIndexRepository indexes = jcas.getJFSIndexRepository();
Iterator psfIter = indexes.getAnnotationIndex(PSESentenceFeature.type).iterator();
Set<String> seSenSpans_ML = new HashSet<String>();
//get side-effect sentence spans found by ML (livSVM)
while(psfIter.hasNext()) {
PSESentenceFeature psf = (PSESentenceFeature) psfIter.next();
//classify SE sentences based on the trained libSVM model
svm_node[] x = new svm_node[psf.getFeatures().size()];
for(int j=0;j<psf.getFeatures().size();j++)
{
x[j] = new svm_node();
x[j].index = j+1;
String nomFea = psf.getFeatures(j);
x[j].value = Double.parseDouble(feaMap.get(nomFea));
}
double clsLabel; //0:non-SE sentence , 1:SE sentence
clsLabel = svm.svm_predict(model,x);
//add side-effect sentence span to the Set
if(clsLabel==1) {
String span = Integer.toString(psf.getPseSen().getBegin()) + "|"
+ Integer.toString(psf.getPseSen().getEnd());
seSenSpans_ML.add(span);
}
}
//get side-effect sentence spans found by rules (SideEffectAnnotator)
Set<String> seSenSpans_rule = new HashSet<String>();
Iterator seIter = indexes.getAnnotationIndex(SideEffectAnnotation.type).iterator();
while(seIter.hasNext()) {
SideEffectAnnotation se = (SideEffectAnnotation) seIter.next();
String span = Integer.toString(se.getSentence().getBegin()) + "|"
+ Integer.toString(se.getSentence().getEnd());
seSenSpans_rule.add(span);
}
//annotate SE sentences (add them to SESentence)
Set<String> seSenSpans = new HashSet<String>(seSenSpans_ML);
seSenSpans.addAll(seSenSpans_rule); //union of ML and rule
for(String s : seSenSpans) {
String[] stk = s.split("\\|");
int begin = Integer.parseInt(stk[0]);
int end = Integer.parseInt(stk[1]);
SESentence ses = new SESentence(jcas);
ses.setBegin(begin);
ses.setEnd(end);
ses.addToIndexes();
}
}
}