Package opennlp.tools.languagemodel

Class NGramLanguageModel

java.lang.Object

opennlp.tools.ngram.NGramModel

opennlp.tools.languagemodel.NGramLanguageModel

All Implemented Interfaces:

Iterable<StringList>, LanguageModel

public class NGramLanguageModel
extends NGramModel
implements LanguageModel

A LanguageModel based on a NGramModel using Stupid Backoff to get the probabilities of the ngrams.

Constructor Summary

Constructors

Constructor	Description
NGramLanguageModel()	Initializes an NGramLanguageModel with DEFAULT_N.
NGramLanguageModel(int n)	Initializes an NGramLanguageModel with the given n for the ngram size.
NGramLanguageModel(InputStream in)	Initializes a NGramLanguageModel instance via a valid InputStream.
NGramLanguageModel(InputStream in, int n)	Initializes a NGramLanguageModel instance via a valid InputStream.

Method Summary

Modifier and Type	Method	Description
void	add(String... tokens)	Adds further tokens.
double	calculateProbability(String... tokens)	Calculate the probability of a series of tokens (e.g.
String[]	predictNextTokens(String... tokens)	Predict the most probable output sequence of tokens, given an input sequence of tokens.
StringList	predictNextTokens(StringList tokens)	Deprecated.

Methods inherited from class opennlp.tools.ngram.NGramModel

add, add, add, contains, cutoff, equals, getCount, hashCode, iterator, numberOfGrams, remove, serialize, setCount, size, toDictionary, toDictionary, toString

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Constructor Detail

NGramLanguageModel

public NGramLanguageModel()

Initializes an NGramLanguageModel with DEFAULT_N.

NGramLanguageModel

public NGramLanguageModel(int n)

Initializes an NGramLanguageModel with the given n for the ngram size.

Parameters:

n - The size of the ngrams to be used. Must be greater than 0.

Throws:

IllegalArgumentException - Thrown if one of the arguments was invalid.

NGramLanguageModel

public NGramLanguageModel(InputStream in)
                   throws IOException

Initializes a NGramLanguageModel instance via a valid InputStream.

Parameters:

in - The InputStream used for loading the model.

Throws:

IOException - Thrown if IO errors occurred during initialization.

IllegalArgumentException - Thrown if one of the arguments was invalid.

NGramLanguageModel

public NGramLanguageModel(InputStream in,
                          int n)
                   throws IOException

Initializes a NGramLanguageModel instance via a valid InputStream.

Parameters:

in - The InputStream used for loading the model.

n - The size of the ngrams to be used. Must be greater than 0.

Throws:

IOException - Thrown if IO errors occurred during initialization.

IllegalArgumentException - Thrown if one of the arguments was invalid.

Method Detail

add

public void add(String... tokens)

Adds further tokens.

Parameters:

tokens - Text elements to add to the NGramLanguageModel.

calculateProbability

public double calculateProbability(String... tokens)

Description copied from interface: LanguageModel

Calculate the probability of a series of tokens (e.g. a sentence), given a vocabulary.

Specified by:

calculateProbability in interface LanguageModel

Parameters:

tokens - the text tokens to calculate the probability for.

Returns:

the probability of the given text tokens in the vocabulary

predictNextTokens

@Deprecated
public StringList predictNextTokens(StringList tokens)

Deprecated.

Description copied from interface: LanguageModel

Predict the most probable output sequence of tokens, given an input sequence of tokens.

Specified by:

predictNextTokens in interface LanguageModel

Parameters:

tokens - a sequence of tokens.

Returns:

the most probable subsequent token sequence

predictNextTokens

public String[] predictNextTokens(String... tokens)

Description copied from interface: LanguageModel

Predict the most probable output sequence of tokens, given an input sequence of tokens.

Specified by:

predictNextTokens in interface LanguageModel

Parameters:

tokens - a sequence of tokens

Returns:

the most probable subsequent token sequence