Transcript An Introduction to GATE

An Introduction to GATE
Presented
by
Lin Lin
What is GATE?

Stands for General Architecture for Text
Engineering.
 The theory behind GATE is SALE
(Software Architecture for Language
Engineering):
– computer processing of human language
– computer infrastructure for software
development
Who Use GATE?

Scientists performing experiments that
involve processing human language
 Developers developing applications with
language processing components
 Teachers and students of courses about
language and language computation
How GATE can Help?

Specify an architecture, or organizational
structure, for language processing software
 Provide a framework, or class library, that
implements the architecture and can be used
to embed language processing capabilities
in diverse applications
 Provide a development environment built
on top of the framework made up of
convenient graphical tools for developing
components
What are GATE Components?

Reusable software chunks with well defined
interfaces
 Used in Java beans and Microsoft’s .Net
GATE as an architecture

Breaks down to three types of components:
– LanguageResources (LRs)
 represent entities such as lexicons, corpora, or
ontologies;
– ProcessingResources (PRs)
 represent entities that are primarily algorithmic, such as
parsers, generators or ngram modelers;
– VisualResources (VRs)
 represent visualization and editing components that
participate in GUIs.
LRs: Corpora, Documents,
and Annotations

A Corpus in Gate is a Java Set whose
members are Documents.
 Documents are modeled as content plus
annotations plus features.
 Annotations are organized in graphs, which
are modeled as Java sets of Annotation.
Documents Processing in GATE

Document:
– Formats including XML, RTF, email, HTML,
SGML, and plain text.
– Identified and converted into GATE annotation
format.
– Processed by PRs.
– Results stored in a serial data store (based on
Java serialization) or as XML.
Built-in GATE Components

Resources for common LE data structures
and algorithms, including documents,
corpora and various annotation types
 A set of language analysis components for
Information Extraction (e.g. ANNIE)
 A range of data visualization and editing
components
Develop Language
Processing Functionality using
GATE

Programming, or the development of
Language Resources such as grammars that
are used by existing Processing Resources,
or a mixture of both.
 The development environment is used for:
– visualization of the data structures produced
and consumed during processing
– debugging
– performance measurement
CREOLE

A Collection of REusable Objects for
Language Engineering
 The set of resources integrated with GATE
 All the resources are packaged as Java
Archive (or ‘JAR’) files, plus some XML
configuration data.
PRs: ANNIE

A family of Processing Resources for
language analysis included with GATE
 Stands for A Nearly-New Information
Extraction system.
 Using finite state techniques to implement
various tasks: tokenization, semantic
tagging, verb phrase chunking, and so on.
ANNIE IE Modules
ANNIE Components

Tokenizer
 Gazetteer
 Sentence Splitter
 Part of Speech Tagger
– produces a part-of-speech tag as an annotation on each
word or symbol.

Semantic Tagger
 OrthoMatcher Coreference Module
ANNIE Component: Tokenizer

Token Types
– word, number, symbol, punctuation, and
spaceToken.

A tokenizer rule has a left hand side and a
right hand side.
Tokenizer Rule

Operations used on the LHS:
– | (or)
– * (0 or more occurrences)
– ? (0 or 1 occurrences)
– + (1 or more occurrences)

The RHS uses ’;’ as a separator, and has the
following format:
{LHS} > {Annotation type};{attribute1}={valu
e1};...;{attribute n}={value n}
Example Tokenizer Rule

"UPPERCASE_LETTER" "LOWERCASE_LETT
ER"* > Token;orth=upperInitial;kind=word;
– The sequence must begin with an uppercase letter,
followed by zero or more lowercase letters. This
sequence will then be annotated as type “Token”. The
attribute “orth” (orthography) has the value
“upperInitial”; the attribute “kind” has the value
“word”.
ANNIE Component: Gazetteer

The gazetteer lists used are plain text files,
with one entry per line.
 Each list represents a set of names, such as
names of cities, organizations, days of the
week, etc.
Example Gazetteer List


A small section of the list for units of currency:
……
Ecu
European Currency Units
FFr
Fr
German mark
German marks
New Taiwan dollar
New Taiwan dollars
NT dollar
NT dollars
……
ANNIE Component:
Semantic Tagger

Based on JAPE language, which contains
rules that act on annotations assigned in
earlier phases.
 Produce outputs of annotated entities.
ANNIE Component: Sentence
Splitter

Segments the text into sentences.

This module is required for the tagger.

The splitter uses a gazetteer list of
abbreviations to help distinguish sentencemarking full stops from other kinds.
ANNIE Component: OrthoMatcher

Adds identity relations between named
entities found by the semantic tagger, in
order to perform coreference.

Does not find new named entities, but it
may assign a type to an unclassified proper
name.
Create a New Resource
Write a Java class that implements GATE’s
beans model.
 Compile the class, and any others that it uses,
into a Java Archive (JAR) file.
 Write some XML configuration data for the
new resource.
 Tell GATE the URL of the new JAR and
XML files.

Example: Create a New
Component Called GoldFish

GoldFish:
– Is a processing resource
– Look for all instances of the word “fish” in the
document
– Add an annotation of type “GoldFish”
Example: Create GoldFish
Using BootStrap Wizard
GoldFish: default files created

The default Java code created for the
GoldFish resource looks like:
– GoldFish.java

The default XML configuration for
GoldFish looks like:
– resource.xml
Create an Application with PRs

Applications model a control strategy for the
execution of PRs.
 Currently only pipeline execution is supported.
– Simple pipelines: group a set of PRs together in
order and execute them in turn.
– Corpus pipelines: open each document in the corpus
in turn, set that document as a runtime parameter on
each PR, run all the PRs on the corpus, then close
the document
Additional Facilities

JAPE
– a Java Annotation Patterns Engine, provides
regular-expression based pattern/action rules over
annotations.
– The file “Main.jape” contains a list of the
grammars to be used for for Named Entity
Recognition, in the correct processing order.
– Used in ANNIE.
Additional Facilities

The ‘annotation diff’ tool in the
development environment
– implements performance metrics such as
precision and recall for comparing annotations.

GUK (the GATE Unicode Kit)
– fills in some of the gaps in the JDK’s support
for Unicode.
Embedding ANNIE


Create a stand alone ANNIE extraction
system.
Example code that will embed ANNIE in
an application that takes URLs as inputs
and produces named entities as outputs.
Additional Features

Add support for a new document format
 Create a new annotation schema
 Write your own algorithm to dump results
to file
 Work with Unicode
 Work with Oracle and PostgreSQL
Other VR can be Used in GATE

Ontogazetteer
– makes ontologies “visible” in GATE.

Protégé
– makes use of developed Protégé ontologies in
GATE, and also take advantage of being able to
read different format ontology files in Protégé.
Link to GATE web page

http://gate.ac.uk
 Documentation and download
GATE Demo

GATE graphical development environment
 Do information extraction with ANNIE
 Create and run an application
 .....