edu.stanford.nlp.ie.crf.CRFClassifierNoisyLabel Maven / Gradle / Ivy

Show more of this group Show more artifacts with this name
Show all versions of stanford-corenlp Show documentation

Stanford CoreNLP provides a set of natural language analysis tools which can take raw English language text input and give the base forms of words, their parts of speech, whether they are names of companies, people, etc., normalize dates, times, and numeric quantities, mark up the structure of sentences in terms of phrases and word dependencies, and indicate which noun phrases refer to the same entities. It provides the foundational building blocks for higher level text understanding applications.

There is a newer version: 4.5.7

Show newest version

// CRFClassifier -- a probabilistic (CRF) sequence model, mainly used for NER.
// Copyright (c) 2002-2008 The Board of Trustees of
// The Leland Stanford Junior University. All Rights Reserved.
//
// This program is free software; you can redistribute it and/or
// modify it under the terms of the GNU General Public License
// as published by the Free Software Foundation; either version 2
// of the License, or (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
//
// For more information, bug reports, fixes, contact:
//    Christopher Manning
//    Dept of Computer Science, Gates 1A
//    Stanford CA 94305-9010
//    USA
//    Support/Questions: [email protected]
//    Licensing: [email protected]

package edu.stanford.nlp.ie.crf; 
import edu.stanford.nlp.util.logging.Redwood;

import edu.stanford.nlp.sequences.*;
import edu.stanford.nlp.util.*;

import java.io.*;

/**
 * Subclass of CRFClassifier for modeling noisy label

 * @author Mengqiu Wang
 */
public class CRFClassifierNoisyLabel extends CRFClassifier  {

  /** A logger for this class */
  private static Redwood.RedwoodChannels log = Redwood.channels(CRFClassifierNoisyLabel.class);

  protected double[][] errorMatrix;

  public CRFClassifierNoisyLabel(SeqClassifierFlags flags) {
    super(flags);
  }

  static double[][] readErrorMatrix(String fileName, Index tagIndex, boolean useLogProb) {
    int numTags = tagIndex.size();
    int matrixSize = numTags;

    String[] matrixLines = new String[matrixSize];
    try {
      BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(new File(fileName))));
      String line = null;
      int lineCount = 0;
      while ((line = br.readLine()) != null) {
        line = line.trim();
        matrixLines[lineCount] = line;
        lineCount++;
      }
    } catch (Exception ex) {
      ex.printStackTrace();
      System.exit(-1);
    }

    double[][] matrix = parseMatrix(matrixLines, tagIndex, matrixSize, false, useLogProb);

    log.info("Error Matrix P(Observed|Truth): ");
    log.info(ArrayUtils.toString(matrix));

    return matrix;
  }

  protected CRFLogConditionalObjectiveFunction getObjectiveFunction(int[][][][] data, int[][] labels) {
    if (errorMatrix == null) {
      if (flags.errorMatrix != null ) {
        if (tagIndex == null) {
          loadTagIndex();
        }
        errorMatrix = readErrorMatrix(flags.errorMatrix, tagIndex, true);
      }
    }
    return new CRFLogConditionalObjectiveFunctionNoisyLabel(data, labels, windowSize, classIndex,
      labelIndices, map, flags.priorType, flags.backgroundSymbol, flags.sigma, null, flags.multiThreadGrad, errorMatrix);
  }
} // end class CRFClassifier