edu.illinois.cs.cogcomp.pos

Class MikheevLearner

java.lang.Object

edu.illinois.cs.cogcomp.lbjava.classify.Classifier

edu.illinois.cs.cogcomp.lbjava.learn.Learner

edu.illinois.cs.cogcomp.pos.POSBaselineLearner

edu.illinois.cs.cogcomp.pos.MikheevLearner

All Implemented Interfaces:

Serializable, Cloneable

Direct Known Subclasses:

MikheevTable

public class MikheevLearner
extends POSBaselineLearner

A poor man's Mikheev rules implementation, in that there is no morphology. Only rules involving all-letter suffixes of lengths 3 and 4 of words whose length is at least 2 greater than the suffix are extracted. The learned rules are divided into three categories: words that are capitalized and appear first in the sentence, words that are capitalized and don't appear first in the sentence, and uncapitalized, unhyphenated words. Uncapitalized, hyphenated words are hard coded to a rule that produces the set { NN, JJ }.

Author:

Nick Rizzolo

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	MikheevLearner.Parameters Empty class, since there are no parameters

Field Summary

Fields

Modifier and Type	Field and Description
protected HashMap<String,TreeMap<String,Integer>>	firstCapitalized A map for capitalized words appearing first in the sentence.
protected HashMap<String,TreeMap<String,Integer>>	notFirstCapitalized A map for capitalized words not appearing first in the sentence.

Fields inherited from class edu.illinois.cs.cogcomp.pos.POSBaselineLearner

table

Fields inherited from class edu.illinois.cs.cogcomp.lbjava.learn.Learner

candidates, encoding, extractor, labeler, labelLexicon, lcFilePath, lexFilePath, lexicon, lossFlag, predictions, readLexiconOnDemand

Fields inherited from class edu.illinois.cs.cogcomp.lbjava.classify.Classifier

containingPackage, name

Constructor Summary

Constructors

Constructor and Description
MikheevLearner() Default constructor; sets the name to the empty string.
MikheevLearner(MikheevLearner.Parameters p) Does nothing, as there are not parameters.
MikheevLearner(String n) Constructor setting the name of the classifier.

Method Summary

Modifier and Type	Method and Description
Set<String>	allowableTags(Word word) Returns the set of tags that the given word's suffix has been observed with, or a reasonable default if the suffix has never been observed.
void	doneLearning() Runs after all learning is complete.
edu.illinois.cs.cogcomp.lbjava.learn.Learner	emptyClone() Returns a new, emtpy learner into which all of the parameters that control the behavior of the algorithm have been copied.
void	forget() Clears out the table to start fresh.
void	learn(Object example) Trains the learning algorithm given an object as an example.
void	prune(HashMap<String,TreeMap<String,Integer>> table) Prunes the specified table.
void	read(ExceptionlessInputStream in) Reads the binary representation of a learner with this object's run-time type, overwriting any and all learned or manually specified parameters as well as the label lexicon but without modifying the feature lexicon.
void	write(ExceptionlessOutputStream out) Writes the learned function's internal representation in binary form.
void	write(PrintStream out) Writes the algorithm's internal representation as text.

Methods inherited from class edu.illinois.cs.cogcomp.pos.POSBaselineLearner

allowableTags, classify, classify, computePrediction, discreteValue, featureValue, getInputType, learn, looksLikeNumber, observed, observedCount, read, scores, write, write

Methods inherited from class edu.illinois.cs.cogcomp.lbjava.learn.Learner

classify, classify, classify, clone, countFeatures, createPrediction, createPrediction, demandLexicon, discreteValue, discreteValue, doneWithRound, featureValue, featureValue, getCurrentLexicon, getExampleArray, getExampleArray, getExtractor, getLabeler, getLabelLexicon, getLexicon, getLexiconDiscardCounts, getLexiconLocation, getModelLocation, getParameters, getPrunedLexiconSize, initialize, learn, learn, learn, read, readLabelLexicon, readLearner, readLearner, readLearner, readLearner, readLearner, readLearner, readLexicon, readLexicon, readLexiconOnDemand, readLexiconOnDemand, readModel, readModel, readParameters, realValue, realValue, realValue, save, saveLexicon, saveModel, scores, scores, scoresAugmented, setCandidates, setEncoding, setExtractor, setLabeler, setLabelLexicon, setLexicon, setLexiconLocation, setLexiconLocation, setLossFlag, setModelLocation, setModelLocation, setParameters, setReadLexiconOnDemand, unclone, unsetLossFlag, write, writeLexicon, writeModel, writeParameters

Methods inherited from class edu.illinois.cs.cogcomp.lbjava.classify.Classifier

allowableValues, classify, discreteValueArray, getCompositeChildren, getOutputType, realValueArray, test, toString, valueIndexOf

Methods inherited from class java.lang.Object

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

Field Detail

firstCapitalized

protected HashMap<String,TreeMap<String,Integer>> firstCapitalized

A map for capitalized words appearing first in the sentence.

notFirstCapitalized

protected HashMap<String,TreeMap<String,Integer>> notFirstCapitalized

A map for capitalized words not appearing first in the sentence.

Constructor Detail

MikheevLearner

public MikheevLearner()

Default constructor; sets the name to the empty string.

MikheevLearner

public MikheevLearner(String n)

Constructor setting the name of the classifier.

Parameters:

n - The name of the classifier.

MikheevLearner

public MikheevLearner(MikheevLearner.Parameters p)

Does nothing, as there are not parameters.

Method Detail

emptyClone

public edu.illinois.cs.cogcomp.lbjava.learn.Learner emptyClone()

Returns a new, emtpy learner into which all of the parameters that control the behavior of the algorithm have been copied. Here, "emtpy" means no learning has taken place.

Overrides:

emptyClone in class POSBaselineLearner

learn

public void learn(Object example)

Trains the learning algorithm given an object as an example.

Overrides:

learn in class POSBaselineLearner

Parameters:

example - An example of the desired learned classifier's behavior.

doneLearning

public void doneLearning()

Runs after all learning is complete.

Overrides:

doneLearning in class edu.illinois.cs.cogcomp.lbjava.learn.Learner

forget

public void forget()

Clears out the table to start fresh.

Overrides:

forget in class POSBaselineLearner

prune

public void prune(HashMap<String,TreeMap<String,Integer>> table)

Prunes the specified table.

Parameters:

table - The table.

allowableTags

public Set<String> allowableTags(Word word)

Returns the set of tags that the given word's suffix has been observed with, or a reasonable default if the suffix has never been observed.

Parameters:

word - The word.

Returns:

The set of tags observed in association with the given word's suffix, or a reasonable default if the suffix has never been observed.

write

public void write(PrintStream out)

Writes the algorithm's internal representation as text.

Overrides:

write in class POSBaselineLearner

Parameters:

out - The output stream.

write

public void write(ExceptionlessOutputStream out)

Writes the learned function's internal representation in binary form.

Overrides:

write in class POSBaselineLearner

Parameters:

out - The output stream.

read

public void read(ExceptionlessInputStream in)

Reads the binary representation of a learner with this object's run-time type, overwriting any and all learned or manually specified parameters as well as the label lexicon but without modifying the feature lexicon.

Overrides:

read in class POSBaselineLearner

Parameters:

in - The input stream.