org.apache.uima.examples.tagger
Class Viterbi

java.lang.Object
  org.apache.uima.examples.tagger.Viterbi

public class Viterbi
extends Object

Viterbi Algorithm: Given a model and a sequence of observations, what is the most likely sequence of states in the model that produces the observations?

Constructor Summary

Viterbi()

Method Summary

static Map<String,List>	init_probs(Map<String,Double> pos_s)
static List	process(int N, List<String> sentence, Map<String,Map<String,Double>> suffix_tree, Map<String,Map<String,Double>> suffix_tree_cap, Map<NGram,Double> transition_probs, Map<String,Map<String,Double>> word_probs, double[] lambdas2, double[] lambdas3, double theta)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Viterbi

public Viterbi()

Method Detail

init_probs

public static Map<String,List> init_probs(Map<String,Double> pos_s)

process

public static List process(int N,
                           List<String> sentence,
                           Map<String,Map<String,Double>> suffix_tree,
                           Map<String,Map<String,Double>> suffix_tree_cap,
                           Map<NGram,Double> transition_probs,
                           Map<String,Map<String,Double>> word_probs,
                           double[] lambdas2,
                           double[] lambdas3,
                           double theta)

Parameters:

N -

sentence -

suffix_tree -

suffix_tree_cap -

transition_probs -

word_probs -

lambdas2 -

lambdas3 -

theta -

Returns:

the viterbi path