Example usage for opennlp.tools.doccat DocumentCategorizerME train

Introduction

In this page you can find the example usage for opennlp.tools.doccat DocumentCategorizerME train.

Prototype

public static DoccatModel train(String languageCode, ObjectStream<DocumentSample> samples,
            TrainingParameters mlParams, DoccatFactory factory) throws IOException 

Source Link

Usage

From source file:io.learningbox.controller.APIController.java

@RequestMapping(value = "/categorize/{area}", method = RequestMethod.POST)
public SortedMap<Double, Set<String>> categorize(@PathVariable final String area, @RequestBody String input)
        throws IOException {
    List<LearningSet> l = repository.findByArea(area);
    final Iterator<LearningSet> sets = l.iterator();

    ObjectStream<DocumentSample> stream = new ObjectStream<DocumentSample>() {

        @Override/*  w  w  w  .j a v  a 2 s . c  om*/
        public DocumentSample read() throws IOException {
            if (sets.hasNext()) {
                LearningSet s = sets.next();

                return new DocumentSample(s.getCategory(), s.getText());
            }
            return null;
        }

        @Override
        public void reset() throws IOException, UnsupportedOperationException {
            throw new UnsupportedOperationException();
        }

        @Override
        public void close() throws IOException {
            //Do nothing
        }
    };

    TrainingParameters trainingParameters = TrainingParameters.defaultParams();
    trainingParameters.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(1000));
    trainingParameters.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(1));

    DoccatModel model = DocumentCategorizerME.train("en", stream, trainingParameters, new DoccatFactory());
    DocumentCategorizerME myCategorizer = new DocumentCategorizerME(model);
    return myCategorizer.sortedScoreMap(input);
}

---