mark.nlp.features
Class TwoLevelOrderer

java.lang.Object
  |
  +--mark.nlp.features.ClassificationOrderer
        |
        +--mark.nlp.features.TwoLevelOrderer

All Implemented Interfaces:

Orderer

Direct Known Subclasses:

Method0, Method3

public abstract class TwoLevelOrderer

extends ClassificationOrderer

A ClassificationOrderer that orders from the corpus counter and the category counters. In other words, it needs to know about the number of occurrences of each potential feature within the corpus, within each category, and within each document.

Field Summary

Fields inherited from class mark.nlp.features.ClassificationOrderer

fCatCounters, fCounter, fPotentialFeatures

Constructor Summary

TwoLevelOrderer()

Method Summary

protected abstract double	combine(int wordIndex, double simpleScore, double[] catScores)
protected java.util.ArrayList	computeScores() Subclasses implement the following function which returns a Score for each potential feature in the set of potential features.

Methods inherited from class mark.nlp.features.ClassificationOrderer

order, setParameter

Methods inherited from class java.lang.Object

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

Constructor Detail

TwoLevelOrderer

public TwoLevelOrderer()

Method Detail

combine

protected abstract double combine(int wordIndex,
                                  double simpleScore,
                                  double[] catScores)

computeScores

protected java.util.ArrayList computeScores()
                                     throws java.lang.Exception

Description copied from class: ClassificationOrderer

Subclasses implement the following function which returns a Score for each potential feature in the set of potential features.

Specified by:

computeScores in class ClassificationOrderer

Returns:

an array list of Scores. The array list is parallel to fVocabulary.

java.lang.Exception