Class NGramFeatureGenerator

Namespace
NOpenNLP.Tools.Doccat
Assembly
NOpenNLP.Tools.dll

Generates ngram features for a document. n-gram IFeatureGenerator

public class NGramFeatureGenerator : IFeatureGenerator
Inheritance
NGramFeatureGenerator
Implements
Inherited Members

Constructors

NGramFeatureGenerator()

Default constructor for Bi grams

public NGramFeatureGenerator()

NGramFeatureGenerator(int, int)

Constructor for ngrams.

public NGramFeatureGenerator(int minGram, int maxGram)

Parameters

minGram int

minGram value - which means minimum words in ngram features

maxGram int

maxGram value - which means maximum words in ngram features

Exceptions

InvalidFormatException

Methods

ExtractFeatures(string[], IDictionary<string, object>)

Extract ngram features from given text fragments

public virtual ICollection<string> ExtractFeatures(string[] text, IDictionary<string, object> extraInfo)

Parameters

text string[]

the text fragments to extract features from

extraInfo IDictionary<string, object>

optional extra information

Returns

ICollection<string>

a collection of n gram features