Java Feature.createName方法代码示例

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * This feature function serves up character n-grams based on StringValued features. For example,
 * if you wanted trigram suffixes (e.g. 'ion' of 'emotion') for words that are of length 7 or more
 * you could call the constructor with the following:
 * CharacterNGramFeatureFunction(Orientation.RIGHT_TO_LEFT, 0, 3, 7, false)
 *
 * @param featureName
 *          a user-specified name for the feature function, to be included in all feature names.
 * @param orientation
 *          must be one of LEFT_TO_RIGHT or RIGHT_TO_LEFT. The orientation determines whether
 *          index 0 corresponds to the first character of the string value or the last. The
 *          orientation does not affect the ordering of the characters in the n-gram which are
 *          always returned in left-to-right order.
 * @param start
 *          the start of the n-gram (typically 0 for both orientations)
 * @param end
 *          the end of the n-gram (typically n for both orientations)
 * @param minimumValueLength
 *          This parameter allows you to skip string values that are too short. It must be greater
 *          than or equal to end.
 * @param lowerCase
 *          if true than the n-gram used as the feature value will be lowercased.
 */
public LTCharacterNgramFeatureFunction(
    String featureName,
    Orientation orientation,
    int start,
    int end,
    int minimumValueLength,
    boolean lowerCase) {
  name = Feature.createName(
      "NGram",
      orientation == Orientation.RIGHT_TO_LEFT ? "Right" : "Left",
      String.valueOf(start),
      String.valueOf(end),
      String.valueOf(minimumValueLength),
      lowerCase ? "lower" : null,
      featureName);
  if (minimumValueLength < end) {
    throw new IllegalArgumentException(
        "minimumValueLength must be greater than or equal to the parameter end.");
  }
  this.orientation = orientation;
  this.start = start;
  this.end = end;
  this.minimumValueLength = minimumValueLength;
  this.lowerCase = lowerCase;
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * @return will return an empty list if the value of the feature is not a StringValue or is not as
 *         long as the minimumValueLength.
 */
@Override
public List<Feature> apply(Feature feature) {
  String featureName = Feature.createName(name, feature.getName());
  Object featureValue = feature.getValue();
  if (featureValue == null || !(featureValue instanceof String)) {
      return Collections.singletonList(new Feature(featureName, "OUT"));
  }

  String value = featureValue.toString();
  if (value == null || value.length() < minimumValueLength) {
      return Collections.singletonList(new Feature(featureName, "OUT"));
  }

  String ngram;
  if (orientation == Orientation.LEFT_TO_RIGHT) {
    ngram = value.substring(start, end);
  } else {
    ngram = value.substring(value.length() - end, value.length() - start);
  }
  if (lowerCase) {
      ngram = ngram.toLowerCase();
  }

  return Collections.singletonList(new Feature(featureName, ngram));
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
public List<NameNumber> encode(Feature feature) throws CleartkEncoderException {
  FeatureCollection fc = (FeatureCollection) feature.getValue();
  List<NameNumber> fves = new ArrayList<NameNumber>();

  if (identifier != null && !identifier.equals(fc.getIdentifier()))
    return Collections.emptyList();

  for (Feature f : fc.getFeatures()) {
    Feature f1 = new Feature(Feature.createName(feature.getName(), f.getName()), f.getValue());
    fves.addAll(subEncoder.encode(f1));
  }

  normalizer.normalize(fves);

  return fves;
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
public List<Feature> extract(JCas jCas, Annotation annotation1, Annotation annotation2) {
  Annotation firstAnnotation, secondAnnotation;

  if (annotation1.getBegin() <= annotation2.getBegin()) {
    firstAnnotation = annotation1;
    secondAnnotation = annotation2;
  } else {
    firstAnnotation = annotation2;
    secondAnnotation = annotation1;
  }

  String featureName = Feature.createName(this.name, "Distance", this.unitClass.getSimpleName());
  int featureValue;
  if (secondAnnotation.getBegin() <= firstAnnotation.getEnd()) {
    featureValue = 0;
  } else {
    List<? extends Annotation> annotations = JCasUtil.selectCovered(
        jCas,
        unitClass,
        firstAnnotation.getEnd(),
        secondAnnotation.getBegin());
    featureValue = annotations.size();
  }

  return Collections.singletonList(new Feature(featureName, featureValue));
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * This feature function serves up character n-grams based on StringValued features. For example,
 * if you wanted trigram suffixes (e.g. 'ion' of 'emotion') for words that are of length 7 or more
 * you could call the constructor with the following:
 * CharacterNGramFeatureFunction(Orientation.RIGHT_TO_LEFT, 0, 3, 7, false)
 * 
 * @param featureName
 *          a user-specified name for the feature function, to be included in all feature names.
 * @param orientation
 *          must be one of LEFT_TO_RIGHT or RIGHT_TO_LEFT. The orientation determines whether
 *          index 0 corresponds to the first character of the string value or the last. The
 *          orientation does not affect the ordering of the characters in the n-gram which are
 *          always returned in left-to-right order.
 * @param start
 *          the start of the n-gram (typically 0 for both orientations)
 * @param end
 *          the end of the n-gram (typically n for both orientations)
 * @param minimumValueLength
 *          This parameter allows you to skip string values that are too short. It must be greater
 *          than or equal to end.
 * @param lowerCase
 *          if true than the n-gram used as the feature value will be lowercased.
 */
public CharacterNgramFeatureFunction(
    String featureName,
    Orientation orientation,
    int start,
    int end,
    int minimumValueLength,
    boolean lowerCase) {
  name = Feature.createName(
      "NGram",
      orientation == Orientation.RIGHT_TO_LEFT ? "Right" : "Left",
      String.valueOf(start),
      String.valueOf(end),
      String.valueOf(minimumValueLength),
      lowerCase ? "lower" : null,
      featureName);
  if (minimumValueLength < end) {
    throw new IllegalArgumentException(
        "minimumValueLength must be greater than or equal to the parameter end.");
  }
  this.orientation = orientation;
  this.start = start;
  this.end = end;
  this.minimumValueLength = minimumValueLength;
  this.lowerCase = lowerCase;
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
@Override
public List<Feature> apply(Feature feature) {
  String featureName = Feature.createName(DEFAULT_NAME, feature.getName());
  Object featureValue = feature.getValue();
  if (featureValue == null)
    return Collections.emptyList();
  else if (featureValue instanceof String) {
    String value = featureValue.toString();
    if (value == null || value.length() == 0)
      return Collections.emptyList();
    if (HyphenUtil.containsHyphen(value))
      return Collections.singletonList(new Feature(featureName, CONTAINS_HYPHEN));
  }
  return Collections.emptyList();
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
@Override
public List<Feature> apply(Feature feature) {
  String featureName = Feature.createName(getFeatureName(), feature.getName());
  Object featureValue = feature.getValue();
  if (featureValue == null)
    return Collections.emptyList();
  else if (featureValue instanceof String) {
    String text = featureValue.toString();
    StringBuilder builder = new StringBuilder();
    String lastType = null;
    boolean multipleRepeats = false;
    for (int i = 0; i < text.length(); i += 1) {
      char c = text.charAt(i);
      String type = classifyChar(c);
      switch (this.patternType) {
        case ONE_PER_CHAR:
          builder.append(type);
          break;
        case REPEATS_MERGED:
          if (!type.equals(lastType)) {
            builder.append(type);
          }
          break;
        case REPEATS_AS_KLEENE_PLUS:
          if (!type.equals(lastType)) {
            builder.append(type);
            multipleRepeats = false;
          } else if (!multipleRepeats) {
            builder.append('+');
            multipleRepeats = true;
          }
      }
      lastType = type;
    }
    return Collections.singletonList(new Feature(featureName, builder.toString()));
  }
  return Collections.emptyList();
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
public List<NameNumber> encode(Feature feature) {
  List<NameNumber> fves = new ArrayList<NameNumber>();
  Counts frequencies = (Counts) feature.getValue();

  String prefix = frequencies.getFeatureName();
  for (Object key : frequencies.getValues()) {
    String name = Feature.createName(prefix, key.toString());
    NameNumber fve = new NameNumber(name, frequencies.getCount(key));
    fves.add(fve);
  }

  normalizer.normalize(fves);

  return fves;
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
public List<Feature> extract(JCas jCas, Annotation annotation1, Annotation annotation2) {
  String featureName = Feature.createName(this.name, "DDistance", this.unitClass.getSimpleName());

  Annotation firstAnnotation, secondAnnotation;
  boolean negate = false;
  if (annotation1.getBegin() <= annotation2.getBegin()) {
    firstAnnotation = annotation1;
    secondAnnotation = annotation2;
  } else {
    firstAnnotation = annotation2;
    secondAnnotation = annotation1;
    negate = true;
  }

  int featureValue = 0;

  if (AnnotationUtil.overlaps(annotation1, annotation2)) {
    featureValue = 0;
  } else {
    List<? extends Annotation> annotations = JCasUtil.selectCovered(
        jCas,
        unitClass,
        firstAnnotation.getEnd(),
        secondAnnotation.getBegin());
    featureValue = annotations.size() + 1;
  }
  if (negate)
    featureValue = -featureValue;

  return Collections.singletonList(new Feature(featureName, featureValue));
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
public Context_ImplBase(int begin, int end) {
  if (begin > end) {
    String message = "expected begin < end, found begin=%d end=%d";
    throw new IllegalArgumentException(String.format(message, begin, end));
  }
  this.begin = begin;
  this.end = end;
  this.name = Feature.createName(
      this.getClass().getSimpleName(),
      String.valueOf(this.begin),
      String.valueOf(this.end));
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * Constructs a {@link Context} which converts the features extracted by the argument contexts
 * into a bag of features where all features have the same name.
 * 
 * @param contexts
 *          The contexts which should be combined into a bag.
 */
public Bag(Context... contexts) {
  this.contexts = contexts;
  String[] names = new String[contexts.length + 1];
  names[0] = "Bag";
  for (int i = 1; i < names.length; ++i) {
    names[i] = contexts[i - 1].getName();
  }
  this.name = Feature.createName(names);
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * Constructs a {@link Context} which converts the features extracted by the argument contexts
 * into a bag of count features.
 * 
 * @param contexts
 *          The contexts which should be combined into a bag.
 */
public Count(Context... contexts) {
  this.contexts = contexts;
  String[] names = new String[contexts.length + 1];
  names[0] = "Count";
  for (int i = 1; i < names.length; ++i) {
    names[i] = contexts[i - 1].getName();
  }
  this.name = Feature.createName(names);
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * If the value of the feature is a StringValue and is determined to be one of ALL_UPPERCASE,
 * ALL_LOWERCASE, INITIAL_UPPERCASE, or MIXED_CASE, then a new feature containing one of those
 * four values is returned. If the value of the feature cannot be characterized by one of these
 * four values, then the empty list is returned (e.g. the value is an empty string, contains only
 * white space, or contains only digits, etc.)
 * 
 * <P>
 * This method was inspired by CapitalizationTypeTagger.py written by Steven Bethard.
 * 
 * @return a feature that has a value that is one of ALL_UPPERCASE, ALL_LOWERCASE,
 *         INITIAL_UPPERCASE, or MIXED_CASE. Otherwise the empty list is returned.
 */
public List<Feature> apply(Feature feature) {
  String featureName = Feature.createName(DEFAULT_NAME, feature.getName());
  Object featureValue = feature.getValue();
  if (featureValue == null)
    return Collections.emptyList();
  else if (featureValue instanceof String) {
    String value = featureValue.toString();
    if (value == null || value.length() == 0)
      return Collections.emptyList();

    String lowerCaseValue = value.toLowerCase();
    String upperCaseValue = value.toUpperCase();
    if (lowerCaseValue.equals(upperCaseValue))
      return Collections.emptyList();

    if (value.equals(value.toLowerCase())) {
      return Collections.singletonList(new Feature(
          featureName,
          CapitalType.ALL_LOWERCASE.toString()));
    } else if (value.equals(value.toUpperCase())) {
      return Collections.singletonList(new Feature(
          featureName,
          CapitalType.ALL_UPPERCASE.toString()));
    }

    if (CaseUtil.isInitialUppercase(value)) {
      return Collections.singletonList(new Feature(
          featureName,
          CapitalType.INITIAL_UPPERCASE.toString()));
    }

    return Collections.singletonList(new Feature(featureName, CapitalType.MIXED_CASE.toString()));
  } else
    return Collections.emptyList();
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * If the value of the feature is a StringValue and is determined to be one of DIGITS,
 * YEAR_DIGITS, ALPHANUMERIC, SOME_DIGITS, or ROMAN_NUMERAL, then a feature containing one of
 * those five values is returned. If the value of the feature cannot be characterized by one of
 * these five values, then an empty list is returned (e.g. the value is an empty string, contains
 * only white space, or contains only letters, etc.)
 * 
 * <p>
 * This method draws heavily from NumericTypeTagger.py written by Steven Bethard. That code
 * credits <a href="http://diveintopython.org/unit_testing/stage_5.html">Dive Into Python</a> for
 * the regular expression for matching roman numerals.
 * 
 * @return a feature that has a value that is one of DIGITS, YEAR_DIGITS, ALPHANUMERIC,
 *         SOME_DIGITS, or ROMAN_NUMERAL. Otherwise an empty list is returned.
 */

@Override
public List<Feature> apply(Feature feature) {
  String featureName = Feature.createName(DEFAULT_NAME, feature.getName());
  Object featureValue = feature.getValue();
  if (featureValue == null)
    return Collections.emptyList();
  else if (featureValue instanceof String) {
    String value = featureValue.toString();
    if (value == null || value.length() == 0)
      return Collections.emptyList();

    if (NumericTypeUtil.isDigits(value)) {
      if (yearDigitsPattern.matcher(value).matches()) {
        return Collections.singletonList(new Feature(
            featureName,
            NumericType.YEAR_DIGITS.toString()));
      } else
        return Collections.singletonList(new Feature(featureName, NumericType.DIGITS.toString()));
    } else if (NumericTypeUtil.containsDigits(value)) {
      if (alphanumericPattern.matcher(value).matches() && someLetters.matcher(value).find()) {
        return Collections.singletonList(new Feature(
            featureName,
            NumericType.ALPHANUMERIC.toString()));
      } else
        return Collections.singletonList(new Feature(
            featureName,
            NumericType.SOME_DIGITS.toString()));
    } else if (romanNumeralPattern.matcher(value).matches()) {
      return Collections.singletonList(new Feature(
          featureName,
          NumericType.ROMAN_NUMERAL.toString()));
    }
  }
  return Collections.emptyList();
}
 

import org.cleartk.ml.Feature; //导入方法依赖的package包/类
/**
 * If the value of the feature is a StringValue and is determined to be one of ALL_UPPERCASE,
 * ALL_LOWERCASE, INITIAL_UPPERCASE, or MIXED_CASE, then a new feature containing one of those
 * four values is returned. If the value of the feature cannot be characterized by one of these
 * four values, then the empty list is returned (e.g. the value is an empty string, contains
 * only white space, or contains only digits, etc.)
 *
 * <P>
 * This method was inspired by CapitalizationTypeTagger.py written by Steven Bethard.
 *
 * @return a feature that has a value that is one of ALL_UPPERCASE, ALL_LOWERCASE,
 *         INITIAL_UPPERCASE, or MIXED_CASE. Otherwise the empty list is returned.
 */
@Override
public List<Feature> apply(Feature feature)
{
    String featureName = Feature.createName(DEFAULT_NAME, feature.getName());
    Object featureValue = feature.getValue();
    if (featureValue == null) {
        return Collections.singletonList(new Feature(featureName, CapitalType.INVALID
                .toString()));
    }
    else if (featureValue instanceof String) {
        String value = featureValue.toString();
        if (value == null || value.length() == 0) {
            return Collections.singletonList(new Feature(featureName, CapitalType.INVALID
                    .toString()));
        }

        String lowerCaseValue = value.toLowerCase();
        String upperCaseValue = value.toUpperCase();
        if (lowerCaseValue.equals(upperCaseValue)) {
            return Collections.singletonList(new Feature(featureName, CapitalType.INVALID
                    .toString()));
        }

        if (value.equals(value.toLowerCase())) {
            return Collections.singletonList(new Feature(featureName, CapitalType.ALL_LOWERCASE
                    .toString()));
        }
        else if (value.equals(value.toUpperCase())) {
            return Collections.singletonList(new Feature(featureName, CapitalType.ALL_UPPERCASE
                    .toString()));
        }

        if (CaseUtil.isInitialUppercase(value)) {
            return Collections.singletonList(new Feature(featureName,
                    CapitalType.INITIAL_UPPERCASE.toString()));
        }

        return Collections.singletonList(new Feature(featureName, CapitalType.MIXED_CASE
                .toString()));
    }
    else {
        return Collections.singletonList(new Feature(featureName, CapitalType.INVALID
                .toString()));
    }
}