org.openimaj.text.nlp.sentiment.model.classifier

Class GeneralSentimentFeatureExtractor

java.lang.Object

org.openimaj.text.nlp.sentiment.model.classifier.GeneralSentimentFeatureExtractor

All Implemented Interfaces:

FeatureExtractor<DoubleFV,List<String>>

public class GeneralSentimentFeatureExtractor
extends Object
implements FeatureExtractor<DoubleFV,List<String>>

FeatureExtractor that is suitable for NaiveBayesAnnotator. Should be initialized with training corpus of the machine learning AbstractAnnotator you are using.

Author:

Laurence Willmore (lgw1e10@ecs.soton.ac.uk)

Constructor Summary

Constructors

Constructor and Description
GeneralSentimentFeatureExtractor() Blank constructor.
GeneralSentimentFeatureExtractor(List<List<String>> domainVocabularyCorpus) Construct with the training set.

Method Summary

Modifier and Type	Method and Description
DoubleFV	extractFeature(List<String> tokens) Extract features from an object and return them.
void	initialize(List<List<String>> domainVocabularyCorpus) Allows a new vocabulary to be constructed from a new corpus.

Methods inherited from class java.lang.Object

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

Constructor Detail

GeneralSentimentFeatureExtractor

public GeneralSentimentFeatureExtractor(List<List<String>> domainVocabularyCorpus)

Construct with the training set. This is required to build a vocabulary.

Parameters:

domainVocabularyCorpus - list of tokenised corpus documents.

GeneralSentimentFeatureExtractor

public GeneralSentimentFeatureExtractor()

Blank constructor. Will require initialize to be called at a later stage.

Method Detail

initialize

public void initialize(List<List<String>> domainVocabularyCorpus)

Allows a new vocabulary to be constructed from a new corpus.

Parameters:

domainVocabularyCorpus - list of tokenised corpus documents.

extractFeature

public DoubleFV extractFeature(List<String> tokens)

Description copied from interface: FeatureExtractor

Extract features from an object and return them.

Specified by:

extractFeature in interface FeatureExtractor<DoubleFV,List<String>>

Parameters:

tokens - the object to extract from

Returns:

the extracted feature