Java MorphoFeatureSpecification类代码示例

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
@Override
public Tree transformTree(Tree t, Tree root) {

  String baseCat = t.value();
  StringBuilder newCategory = new StringBuilder();

  //Add manual state splits
  for (Pair<TregexPattern,Function<TregexMatcher,String>> e : activeAnnotations) {
    TregexMatcher m = e.first().matcher(root);
    if (m.matchesAt(t))
      newCategory.append(e.second().apply(m));
  }

  //Add morphosyntactic features if this is a POS tag
  if(t.isPreTerminal() && tagSpec != null) {
    if( !(t.firstChild().label() instanceof CoreLabel) || ((CoreLabel) t.firstChild().label()).originalText() == null )
      throw new RuntimeException(String.format("%s: Term lacks morpho analysis: %s",this.getClass().getName(),t.toString()));

    String morphoStr = ((CoreLabel) t.firstChild().label()).originalText();
    Pair<String,String> lemmaMorph = MorphoFeatureSpecification.splitMorphString("", morphoStr);
    MorphoFeatures feats = tagSpec.strToFeatures(lemmaMorph.second());
    baseCat = feats.getTag(baseCat);
  }

  //Update the label(s)
  String newCat = baseCat + newCategory.toString();
  t.setValue(newCat);
  if (t.isPreTerminal() && t.label() instanceof HasTag)
    ((HasTag) t.label()).setTag(newCat);

  return t;
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
/**
 * Convert token to a sequence of datums and add to iobList.
 * 
 * @param iobList
 * @param tokenText
 * @param tokenLabel
 * @param lastToken
 * @param charIndex
 * @param applyRewriteRules
 */
private static void tokenToDatums(List<CoreLabel> iobList, String token, TokenType tokType,
        CoreLabel tokenLabel, String lastToken, int charIndex, boolean applyRewriteRules) {
    String lastLabel = ContinuationSymbol;
    String firstLabel = BeginSymbol;
    if (applyRewriteRules) {
        // Apply Arabic-specific re-write rules
        String rawToken = tokenLabel.word();
        String tag = tokenLabel.tag();
        MorphoFeatureSpecification featureSpec = new ArabicMorphoFeatureSpecification();
        featureSpec.activate(MorphoFeatureType.NGEN);
        featureSpec.activate(MorphoFeatureType.NNUM);
        MorphoFeatures features = featureSpec.strToFeatures(tag);

        // Rule #1 : ت --> ة
        if (features.getValue(MorphoFeatureType.NGEN).equals("F")
                && features.getValue(MorphoFeatureType.NNUM).equals("SG") && rawToken.endsWith("ت-")) {
            lastLabel = RewriteTahSymbol;
        }

        // Rule #2 : لل --> ل ال
        if (lastToken.equals("ل") && rawToken.startsWith("-ل")) {
            firstLabel = RewriteTareefSymbol;
        }
    }
    int index = tokenLabel.get(CoreAnnotations.CharacterOffsetBeginAnnotation.class);
    String origToken = tokenLabel.get(CoreAnnotations.OriginalTextAnnotation.class);
    // Create datums and add to iobList
    String firstChar = String.valueOf(token.charAt(0));
    iobList.add(createDatum(firstChar, firstLabel, charIndex++, firstChar,
        String.valueOf(origToken.charAt(0)), index++,
        index, tokenLabel.get(CoreAnnotations.BeforeAnnotation.class)));
    final int numChars = token.length();
    for (int j = 1; j < numChars; ++j) {
        String thisChar = String.valueOf(token.charAt(j));
        String charLabel = (j == numChars - 1) ? lastLabel : ContinuationSymbol;
        iobList.add(createDatum(thisChar, charLabel, charIndex++, thisChar, 
            String.valueOf(origToken.charAt(j)),index++, index, ""));
    }
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
/**
 * Returns a morphological feature specification for words in this language.
 */
@Override
public MorphoFeatureSpecification morphFeatureSpec() {
  return null;
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
public FactoredLexicon(MorphoFeatureSpecification morphoSpec, Index<String> wordIndex, Index<String> tagIndex) {
  super(wordIndex, tagIndex);
  this.morphoSpec = morphoSpec;
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
/**
 * This method should populate wordIndex, tagIndex, and morphIndex.
 */
@Override 
public void train(Collection<Tree> trees, Collection<Tree> rawTrees) {
  double weight = 1.0;
  // Train uw model on words
  uwModelTrainer.train(trees, weight);
  
  final double numTrees = trees.size();
  Iterator<Tree> rawTreesItr = rawTrees == null ? null : rawTrees.iterator();
  Iterator<Tree> treeItr = trees.iterator();
  
  // Train factored lexicon on lemmas and morph tags
  int treeId = 0;
  while (treeItr.hasNext()) {
    Tree tree = treeItr.next();
    // CoreLabels, with morph analysis in the originalText annotation
    List<Label> yield = rawTrees == null ? tree.yield() : rawTreesItr.next().yield();
    // Annotated, binarized tree for the tags (labels are usually CategoryWordTag)
    List<Label> pretermYield = tree.preTerminalYield();

    int yieldLen = yield.size();
    for (int i = 0; i < yieldLen; ++i) {
      String word = yield.get(i).value();
      int wordId = wordIndex.indexOf(word, true); // Don't do anything with words
      String tag = pretermYield.get(i).value();
      int tagId = tagIndex.indexOf(tag, true);

      // Use the word as backup if there is no lemma
      String featureStr = ((CoreLabel) yield.get(i)).originalText();
      Pair<String,String> lemmaMorph = MorphoFeatureSpecification.splitMorphString(word, featureStr);
      String lemma = lemmaMorph.first();
      int lemmaId = wordIndex.indexOf(lemma, true);
      String richMorphTag = lemmaMorph.second();
      String reducedMorphTag = morphoSpec.strToFeatures(richMorphTag).toString().trim();
      reducedMorphTag = reducedMorphTag.length() == 0 ? NO_MORPH_ANALYSIS : reducedMorphTag;
      int morphId = morphIndex.indexOf(reducedMorphTag, true);
      
      // Seen event counts
      wordTag.incrementCount(wordId, tagId);
      lemmaTag.incrementCount(lemmaId, tagId);
      morphTag.incrementCount(morphId, tagId);
      tagCounter.incrementCount(tagId);
      
      // Unseen event counts
      if (treeId > op.trainOptions.fractionBeforeUnseenCounting*numTrees) {
        if (! wordTag.firstKeySet().contains(wordId) || wordTag.getCounter(wordId).totalCount() < 2) {
          wordTagUnseen.incrementCount(tagId);
        }
        if (! lemmaTag.firstKeySet().contains(lemmaId) || lemmaTag.getCounter(lemmaId).totalCount() < 2) {
          lemmaTagUnseen.incrementCount(tagId);
        }
        if (! morphTag.firstKeySet().contains(morphId) || morphTag.getCounter(morphId).totalCount() < 2) {
          morphTagUnseen.incrementCount(tagId);
        }
      }
    }
    ++treeId;

    if (DEBUG && (treeId % 100) == 0) {
      System.err.printf("[%d]",treeId);
    }
    if (DEBUG && (treeId % 10000) == 0) {
      System.err.println();
    }
  }
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
@Override
public MorphoFeatureSpecification morphFeatureSpec() {
  return new ArabicMorphoFeatureSpecification();
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
@Override
public Tree normalizeWholeTree(Tree tree, TreeFactory tf) {
  tree = tree.prune(emptyFilter, tf).spliceOut(aOverAFilter, tf);

  for(Tree t : tree) {
    //Map punctuation tags back like the PTB
    if(t.isPreTerminal()) {
      String posStr = normalizePreterminal(t);
      t.setValue(posStr);
      if(t.label() instanceof HasTag) ((HasTag) t.label()).setTag(posStr);

    } else if(t.isLeaf()) {
      //Strip off morphological analyses and place them in the OriginalTextAnnotation, which is
      //specified by HasContext.
      if(t.value().contains(MorphoFeatureSpecification.MORPHO_MARK)) {
        String[] toks = t.value().split(MorphoFeatureSpecification.MORPHO_MARK);
        if(toks.length != 2)
          System.err.printf("%s: Word contains malformed morph annotation: %s%n",this.getClass().getName(),t.value());

        else if(t.label() instanceof CoreLabel) {
          ((CoreLabel) t.label()).setValue(toks[0].trim().intern());
          ((CoreLabel) t.label()).setWord(toks[0].trim().intern());
          ((CoreLabel) t.label()).setOriginalText(toks[1].trim().intern());
        } else {
          System.err.printf("%s: Cannot store morph analysis in non-CoreLabel: %s%n",this.getClass().getName(),t.label().getClass().getName());
        }
      }
    }
  }

  //Add start symbol so that the root has only one sub-state. Escape any enclosing brackets.
  //If the "tree" consists entirely of enclosing brackets e.g. ((())) then this method
  //will return null. In this case, readers e.g. PennTreeReader will try to read the next tree.
  while(tree != null && (tree.value() == null || tree.value().equals("")) && tree.numChildren() <= 1)
    tree = tree.firstChild();

  //Ensure that the tree has a top-level unary rewrite
  if(tree != null && !tree.value().equals(rootLabel))
    tree = tf.newTreeNode(rootLabel, Collections.singletonList(tree));

  return tree;
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
@Override
public MorphoFeatureSpecification morphFeatureSpec() {
  return new FrenchMorphoFeatureSpecification();
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
/**
 * Returns a morphological feature specification for words in this language.
 */
public MorphoFeatureSpecification morphFeatureSpec() {
  return null;
}
 

import edu.stanford.nlp.international.morph.MorphoFeatureSpecification; //导入依赖的package包/类
/**
 * The morphological feature specification for the language.
 *
 * @return A language-specific MorphoFeatureSpecification
 */
public abstract MorphoFeatureSpecification morphFeatureSpec();