edu.cmu.minorthird.classify.ranking
Class BatchRankingLearner

java.lang.Object
  edu.cmu.minorthird.classify.BatchClassifierLearner
      edu.cmu.minorthird.classify.BatchBinaryClassifierLearner
          edu.cmu.minorthird.classify.ranking.BatchRankingLearner

All Implemented Interfaces:

BinaryClassifierLearner, ClassifierLearner, java.lang.Cloneable

Direct Known Subclasses:

ListNet, RankingBoosted, RankingPerceptron

public abstract class BatchRankingLearner
extends BatchBinaryClassifierLearner

Learns to rank sets of examples, rather than to classify individual examples. Data is presented to a BatchRankingLearner as an ordinary binary dataset. Examples from the same subpopulation are comparable, and should be ranked so that positive examples have a higher score than negative examples.

Author:

William Cohen

Field Summary

Fields inherited from class edu.cmu.minorthird.classify.BatchClassifierLearner

classifier, dataset

Constructor Summary

BatchRankingLearner()

Method Summary

static java.util.Map<java.lang.String,java.util.List<Example>>	listsWithOneExampleEach(java.util.Map<java.lang.String,java.util.List<Example>> rankingLists) Split a Map output by splitIntoRankings into lists that contain exactly one positive example each.
static void	sortByScore(BinaryClassifier c, java.util.List<Example> data) Sort a List of Instances by score according to the classifier.
static java.util.Map<java.lang.String,java.util.List<Example>>	splitIntoRankings(Dataset data) Sort a dataset into 'rankings'.

Methods inherited from class edu.cmu.minorthird.classify.BatchBinaryClassifierLearner

batchTrainBinary, getBinaryClassifier, getSchema, setSchema

Methods inherited from class edu.cmu.minorthird.classify.BatchClassifierLearner

addExample, batchTrain, completeTraining, copy, getClassifier, hasNextQuery, nextQuery, reset, setInstancePool

Methods inherited from class java.lang.Object

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

Methods inherited from interface edu.cmu.minorthird.classify.ClassifierLearner

addExample, completeTraining, copy, getClassifier, hasNextQuery, nextQuery, reset, setInstancePool

Constructor Detail

BatchRankingLearner

public BatchRankingLearner()

Method Detail

splitIntoRankings

public static java.util.Map<java.lang.String,java.util.List<Example>> splitIntoRankings(Dataset data)

Sort a dataset into 'rankings'. Each ranking is a List of Examples such that all positive examples in the list should be ranked above all negative examples. Returns a map so that map.get(key) is an ArrayList of examples.

listsWithOneExampleEach

public static java.util.Map<java.lang.String,java.util.List<Example>> listsWithOneExampleEach(java.util.Map<java.lang.String,java.util.List<Example>> rankingLists)

Split a Map output by splitIntoRankings into lists that contain exactly one positive example each.

sortByScore

public static void sortByScore(BinaryClassifier c,
                               java.util.List<Example> data)

Sort a List of Instances by score according to the classifier.