de.jungblut.ner

Class BasicFeatureExtractor

java.lang.Object

de.jungblut.ner.BasicFeatureExtractor

All Implemented Interfaces:

SequenceFeatureExtractor<String>

public final class BasicFeatureExtractor
extends Object
implements SequenceFeatureExtractor<String>

Basic feature extraction for sequence learning, takes the current word into account and the previous label - as well as the joint version of both. This will implicitly create a dictionary in your features.
This was used to derive names from an english text.

Author:

thomas.jungblut

Constructor Summary

Constructors

Constructor and Description
BasicFeatureExtractor()

Method Summary

Modifier and Type	Method and Description
List<String>	computeFeatures(List<String> words, int previousLabel, int position) Compute a feature for the given sequence (the complete list words).

Methods inherited from class java.lang.Object

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

Constructor Detail

BasicFeatureExtractor

public BasicFeatureExtractor()

Method Detail

computeFeatures

public List<String> computeFeatures(List<String> words,
                                    int previousLabel,
                                    int position)

Description copied from interface: SequenceFeatureExtractor

Compute a feature for the given sequence (the complete list words). Given are the previous label and the current index (position). This method will be called for every index in the list of words.

Specified by:

computeFeatures in interface SequenceFeatureExtractor<String>

Parameters:

words - all words in that sequence.

previousLabel - the previous label.

position - the current position.

Returns:

a set of features for this position.