A B E G I N P S T

A

abner - package abner

B

BIOCREATIVE - Static variable in class abner.Tagger

The tagger trained on the BioCreative corpus.

E

EXTERNAL - Static variable in class abner.Tagger

Indicates a tagger for some externally-trained model.

G

getEntities(String) - Method in class abner.Tagger

Similar to getSegments, but returns all segments in the entire document that correspond to entities (e.g.

getEntities(String, String) - Method in class abner.Tagger

Returns only segments corresponding to the entity provided in the tag argument (do not us "B-" or "I-" prefixes).

getMode() - Method in class abner.Tagger

Return the tagger's mode (NLPBA, BIOCREATIVE, or EXTERNAL)

getSegments(String) - Method in class abner.Tagger

Take an input string (if tokenization is turned on, this string will be tokenized as well) and return a Vector of 2D String arrays, where sentence tokens are segments (not individual words).

getTokenization() - Method in class abner.Tagger

Return the tagger's current tokenization setting.

getWords(String) - Method in class abner.Tagger

Take an input string (if tokenization is turned on, this string will be tokenized as well) and return a Vector of 2D String arrays, where sentence tokens are words stored in result[0][...] and tags are stored in result[1][...].

I

Input2TokenSequence - class abner.Input2TokenSequence.

Input2TokenSequence is a text processing Pipe for the MALLET framework.

Input2TokenSequence(boolean) - Constructor for class abner.Input2TokenSequence

Input2TokenSequence() - Constructor for class abner.Input2TokenSequence

N

NLPBA - Static variable in class abner.Tagger

The tagger trained on the NLPBA corpus.

nextToken() - Method in class abner.Scanner

P

pipe(Instance) - Method in class abner.Input2TokenSequence

S

Scanner - class abner.Scanner.

ABNER's Scanner class implements the finite state machine used in tokenization.

Scanner(Reader) - Constructor for class abner.Scanner

Scanner(InputStream) - Constructor for class abner.Scanner

setTokenization(boolean) - Method in class abner.Tagger

Turn on/off ABNER's built-in tokenization (default is true).

T

Tagger - class abner.Tagger.

This is the interface to the CRF that does named entity tagging.

Tagger() - Constructor for class abner.Tagger

Basic Constructor: Loads the "NLPBA" model by default.

Tagger(int) - Constructor for class abner.Tagger

Advanced constructor: Specify either "NLPBA" or "BioCreative" model.

Tagger(File) - Constructor for class abner.Tagger

External constructor: Load a trained CRF specified by the external model file.

Trainer - class abner.Trainer.

The Trainer class will train a CRF to extract entities from a customized dataset.

Trainer() - Constructor for class abner.Trainer

tagABNER(String) - Method in class abner.Tagger

Takes input text and returns a string of annotated text in the ABNER training format:

tagIOB(String) - Method in class abner.Tagger

Takes input text and returns a string of annotated text in CoNLL-style "IOB" format:

tagSGML(String) - Method in class abner.Tagger

Takes input text and returns a string of annotated text in a generic SGML-style format:

tokenize(String) - Method in class abner.Tagger

Take raw text apply ABNER's built-in tokenization on it.

train(String, String) - Method in class abner.Trainer

Takes input trainFile (format described above), and saves a trained linear-chain CRF on the data using ABNER's default feature set in the corresponding output modelFile.

train(String, String, String[]) - Method in class abner.Trainer

Identical to the other train routine, but the set of tags (e.g.

A B E G I N P S T