dragon.nlp.tool

Interface NER

All Known Implementing Classes:

Annie

public interface NER

Interface of Named Entity Recognizer

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0

Author:

Davis Zhou

Method Summary

Modifier and Type	Method and Description
String	annotate(String doc) The format of an annotated text is XML.
ArrayList	extractEntities(String doc) Extract entities from a given text.
ArrayList	extractFromDoc(Document doc) The return format is the same as the one by the method extractFromSentence.
ArrayList	extractFromSentence(Sentence sent) Difering from the method extractEntites, this method will save an entity in a term and thus all information such as entity type is kept.
void	setAnnotationTypes(String[] arrType) Set the entity types of interest, for example, organization, location, person, etc.

Method Detail

extractFromDoc

ArrayList extractFromDoc(Document doc)

The return format is the same as the one by the method extractFromSentence.

Parameters:

doc - a parsed document

Returns:

a list of extracted entities in their original order in the text.

extractFromSentence

ArrayList extractFromSentence(Sentence sent)

Difering from the method extractEntites, this method will save an entity in a term and thus all information such as entity type is kept. Moreover, duplicated entities will not be merged. Instead, all extracted entities will be returned in their orders in the text.

Parameters:

sent - the sentence for extraction

Returns:

a list of extracted entities in their original order in the text.

extractEntities

ArrayList extractEntities(String doc)

Extract entities from a given text. Each extracted entity will be save in a Token. Duplicated entities will be merged. The frequency of a token is actually the occurrence frequency of the entity in the text. The type of the entity will be ignored.

Parameters:

doc - the text for processing

Returns:

a list of unique entities with their frequency.

annotate

String annotate(String doc)

The format of an annotated text is XML. Each entity in the text will be tagged by a XML node and the attribute of the node will be the type of the entity.

Parameters:

doc - the text for annotation

Returns:

an annotated text

setAnnotationTypes

void setAnnotationTypes(String[] arrType)

Set the entity types of interest, for example, organization, location, person, etc. The interpretation of these types is subject to the implemented named entity recognizers.

Parameters:

arrType - a set of wanted entity types.