Class DefaultLemmatizerContextGenerator

java.lang.Object
opennlp.tools.lemmatizer.DefaultLemmatizerContextGenerator
All Implemented Interfaces:
LemmatizerContextGenerator, BeamSearchContextGenerator<String>

public class DefaultLemmatizerContextGenerator extends Object implements LemmatizerContextGenerator
Simple feature generator for learning statistical lemmatizers.

Features based on Grzegorz ChrupaƂa. 2008. Towards a Machine-Learning Architecture for Lexical Functional Grammar Parsing. PhD dissertation, Dublin City University

  • Constructor Details

    • DefaultLemmatizerContextGenerator

      public DefaultLemmatizerContextGenerator()
  • Method Details

    • getContext

      public String[] getContext(int index, String[] sequence, String[] priorDecisions, Object[] additionalContext)
      Description copied from interface: BeamSearchContextGenerator
      Returns the context for the specified position in the specified sequence (list).
      Specified by:
      getContext in interface BeamSearchContextGenerator<String>
      Parameters:
      index - The index of the sequence.
      sequence - The sequence of items over which the beam search is performed.
      priorDecisions - The sequence of decisions made prior to the context for which this decision is being made.
      additionalContext - Any addition context specific to a class implementing this interface.
      Returns:
      The context for the specified index in the specified sequence.
    • getContext

      public String[] getContext(int index, String[] toks, String[] tags, String[] preds)
      Description copied from interface: LemmatizerContextGenerator
      Returns the contexts for lemmatizing of the specified index.
      Specified by:
      getContext in interface LemmatizerContextGenerator
      Parameters:
      index - The index of the token in the specified toks array for which the context should be constructed.
      toks - The tokens of the sentence. The toString() methods of these objects should return the token text.
      tags - The POS tags for the specified toks.
      preds - The previous decisions made in the tagging of this sequence. Only indices less than i will be examined.
      Returns:
      An array of predictive contexts on which a model basis its decisions.