Package opennlp.tools.doccat

Class NGramFeatureGenerator

java.lang.Object

opennlp.tools.doccat.NGramFeatureGenerator

All Implemented Interfaces:

FeatureGenerator

public class NGramFeatureGenerator
extends Object
implements FeatureGenerator

Generates ngram features for a document.

See Also:

• FeatureGenerator

Constructor Summary

Constructors

Constructor	Description
NGramFeatureGenerator()	Instantiates an NGramFeatureGenerator instance with a Bi grams config.
NGramFeatureGenerator(int minGram, int maxGram)	Instantiates an NGramFeatureGenerator instance with configurable ngram parameters.

Method Summary

Modifier and Type	Method	Description
Collection<String>	extractFeatures(String[] text, Map<String,Object> extraInfo)	Extracts features from given text fragments.

Methods inherited from class java.lang.Object

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

Constructor Details

NGramFeatureGenerator

public NGramFeatureGenerator(int minGram,
 int maxGram)
                      throws InvalidFormatException

Instantiates an NGramFeatureGenerator instance with configurable ngram parameters.

Parameters:

minGram - The minimum words in ngram features.

maxGram - The maximum words in ngram features.

Throws:

InvalidFormatException - Thrown if parameter values are invalid or inconsistent.

NGramFeatureGenerator

public NGramFeatureGenerator()
                      throws InvalidFormatException

Instantiates an NGramFeatureGenerator instance with a Bi grams config.

Throws:

InvalidFormatException - Thrown if parameter values are invalid or inconsistent.

Method Details

extractFeatures

public Collection<String> extractFeatures(String[] text,
 Map<String,Object> extraInfo)

Description copied from interface: FeatureGenerator

Extracts features from given text fragments.

Specified by:

extractFeatures in interface FeatureGenerator

Parameters:

text - The text fragments to extract features from

extraInfo - Optional extra information to be used by the FeatureGenerator.

Returns:

A collection of features.