+ All Categories
Home > Documents > Natural Language Generation

Natural Language Generation

Date post: 19-Jan-2016
Category:
Upload: sawyer
View: 17 times
Download: 2 times
Share this document with a friend
Description:
Natural Language Generation. Ling 571 Fei Xia Week 8: 11/17/05. Outline. NLG basics What is NLG? Example NLG systems Applications of NLG Architecture for NLG Three subtasks of NLG Document planning Microplanning Surface realization. NLG basics. What is NLG?. - PowerPoint PPT Presentation
37
Natural Language Generation Ling 571 Fei Xia Week 8: 11/17/05
Transcript
Page 1: Natural Language Generation

Natural Language Generation

Ling 571

Fei Xia

Week 8: 11/17/05

Page 2: Natural Language Generation

Outline

• NLG basics– What is NLG?– Example NLG systems– Applications of NLG– Architecture for NLG

• Three subtasks of NLG– Document planning– Microplanning– Surface realization

Page 3: Natural Language Generation

NLG basics

Page 4: Natural Language Generation

What is NLG?

• Definition (McDonald 1992): the process of deliberately constructing a NL text in order to meet specified communicative goals.

• Input: non-linguistic representation of info

• Output: text, hypertext, speech

Page 5: Natural Language Generation

NLU vs. NLG

• NLU: ambiguity, ill-formed input

• NLG: choice

Text Text

Meaning

NLU NLG

Page 6: Natural Language Generation

NLG system #1: FoG

• FoG: Forecast Generator

• Input: weather map

• Output: textual weather report in English and French

• Developer: CoGen Tex

• Status: in operational use since 1992

Page 7: Natural Language Generation

NLG system #2: SumTime-Mousam

• FoG: Forecast Generator

• Input: weather data

• Output: textual weather report in English

• Developer: University of Aberdeen

• Status: Used by one company to generate weather forecasts for offshore oil rigs.

• Demo

Page 8: Natural Language Generation

NLG System #3: STOP

• Input: Questionnaire about smoking attitudes, history, beliefs

• Output: a personalized smoking-cessation leaflet

• Developer: University of Aberdeen

• Demo: non-tailored letter, questionnaire, a tailored letter.

• Status: undergoing clinical evaluation

Page 9: Natural Language Generation

Applications of NLG

• Automated document production:– Weather forecasts, letters, ….

• Dialogue systems: e.g., ELIZA

• Text summarization

• Machine translation

• Question Answering

• ….

Page 10: Natural Language Generation

ElIZA system• H: My head hurts• E: Why do you say your head hurts?

• H: My mother hates me.• E: Who else in your family hates you?

• H: *******.• E: Tell me more about it.

• H: What should I do?• E: That’s a tough decision.

• H: ****?• E: Does that question interest you?

Page 11: Natural Language Generation

ELIZA system (cont)

• The famous 1966 computer system by Joseph Weizenbaum, which acted as a therapist

• Named after “Eliza Doolittle” in in Shaw's Pygmalion (“My fair lady”).

• Use pattern matching rules

• Very successful: prompt him to write a book to explain the limits of computers

Page 12: Natural Language Generation

Challenges of NLG

• Decide what to say and how to say it.

• The output has to be well-formed text.– Orthography, morphology, syntax– Reference, word choice, pragmatics

• Example: weather report

Page 13: Natural Language Generation

Component tasks in NLG

• Content determination: what content to express• Document structuring: how to structure the info

to make a coherent text.• Aggregation: combine units in a document plan

tree.• Lexicalization: what words to use?• Referring expression generation: NP or

pronoun?• Structure realization: add markups • Linguistic realization: tense, aspect, voice, …

Page 14: Natural Language Generation

Tasks and modules in NLG

Document

Planning

Content determination

Document structuring

Microplanning Aggregation

Lexicalization

Referring expression generation

Surface realisation

Linguistic realization

Structure Realization

Page 15: Natural Language Generation

A pipelined architecture

Document Planning

Microplanning

Surface realization

Document plan

Text Specification

Text

Communicative goalKnowledge baseGrammar ….

Page 16: Natural Language Generation

Another architecture

Communicative Goal Knowledge base

Discourse Planner

Surface Realizer

NL output

Discourse specification

Page 17: Natural Language Generation

Outline

• NLG basics– What is NLG?– Example NLG systems– Applications of NLG– Architecture for NLG

• Three subtasks of NLG– Document planning– Microplanning– Surface realization

Page 18: Natural Language Generation

Document planning

• Content determination: – What is important?– a domain-dependent expert-system-like task.– Ex: Weather summary:

• This Nov was very dry.• The temperature was lower.

• Document structuring: – Use RST or other discourse theory

Page 19: Natural Language Generation

Document plan tree

Msg1

NucleusSatellite(contrast)

Msg2

NucleusSatellite(Elaboration)

NucleusNucleus(sequence)

Msg3

Page 20: Natural Language Generation

Microplanning

• Aggregation

• Lexicalization

• Referring expression generation

Page 21: Natural Language Generation

Aggregation

• Combinations can be on the basis of – Information content– Possible forms of realization

• Some possibilities:– Conjunction– Ellipsis– Embedding– …

Page 22: Natural Language Generation

Aggregation via conjunction

• Without aggregation:– Light rain fell on the 6th.– Light rain fell on the 8th.

• With Aggregation:– Light rain fell on the 6th and light rain fell on

the 8th. (conjunction)– Light rain fell on the 6th and the 8th.

Page 23: Natural Language Generation

Aggregation via embedding

• Without aggregation: – November had a rainfall of 20mm.– November normally is the wettest month.

• With aggregation:– November, which normally is the wettest

month, had a rainfall of 20mm this year.– Although November is the wettest month, this

November had a rainfall of only 20mm.

Page 24: Natural Language Generation

Aggregation strategies

• Conform to genre conventions and rules, and take account of pragmatic goals:– Ex: making sentences shorter for poor

readers

• Observe structural properties: – Ex: aggregate only messages that are siblings

in the document plan tree.

Page 25: Natural Language Generation

An aggregation rule

Msg1

Nucleus Satellite(contrast)

Msg2

S

S

Conj S

S

Msg2

Msg1

although

Page 26: Natural Language Generation

Lexicalization

• The process of choosing words to communicate the info in messages

• When several lexicalizations are possible, consider:– User knowledge and preference– Consistency with previous usage: sometimes,

it is better to vary lexemes– Pragmatics

Page 27: Natural Language Generation

Examples

• Light rain• A small amount of rain

• It is encouraging that you have many reasons to stop.

• It’s good that you have a lot of reasons to stop.

Page 28: Natural Language Generation

Referring expression generation

• How do we identify specific objects and entities?

• Two cases:– Initial introduction of an object– Subsequence references to an already

mentioned object.

Page 29: Natural Language Generation

Initial reference

A few options:

• Use a full name: John Smith

• Relate to an entity that is already salient– One of Dr. Klein’s patients– The person who came to the clinic yesterday

Page 30: Natural Language Generation

Subsequence reference

Some possibilities:

• Pronouns: He is very determined.

• Definite NPs: This person is very determined.

• Proper names: John is very talented.

Page 31: Natural Language Generation

Choosing referring expressions

• Some suggestions from the literature:– Use a pronoun if it refers to an entity

mentioned in the previous clause, and there is no other entity in the previous clause that this pronoun could refer to.

– Otherwise, use a name (a short one if possible)

– Otherwise, use a definite NP.

Page 32: Natural Language Generation

Choosing referring expressions (cont)

• Considering genre conventions and the context

• Ex: – Nov 2005– November– This month

Page 33: Natural Language Generation

Surface realization

• Goal: to convert text specification into actual text.

• Structural realization: e.g., add html markup

• Linguistics realization:– Insert function words– Choose correct inflection– Order words within a sentence– Add punctuation

Page 34: Natural Language Generation

Linguistics realization

• Systemic Grammar • Functional unification grammar

• Ex: – Input: (:action rain :tense past :time November :degree little)– Output: it rained little in November

Page 35: Natural Language Generation

Summary

• NLG basics– What is NLG?– Example NLG systems– Applications of NLG– Architecture for NLG

• Three subtasks of NLG– Document planning– Microplanning– Surface realization

Page 36: Natural Language Generation

Beyond text generation

• Flat text

• Structured text: itemized lists, section, chapter, ….

• Text and graphics: e.g., picture with caption

• Speech

• ….

Page 37: Natural Language Generation

Resources

• SIGGEN (ACL special interest group for Generation): www.siggen.org

• Book: “Building NLG systems” by Reiter et. al., Cambridge University Press, 2000.

• List of NLG systems:

• Companies:– CoGenTex: www.cogentex.com– ERLI: www.erli.com


Recommended