Natural Language Generation: Discourse Planning

Slides:



Advertisements
Similar presentations
Helmut König, Siemens Medical Solutions
Advertisements

December 2003CSA3050: Natural Language Generation 1 What is Natural Language Generation? When is NLG an Appropriate Technology? NLG System Architectures.
CTS130 Spreadsheet Lesson 20 Data Consolidation. Consolidation is a process in which data from multiple worksheets or workbooks is combined and summarized.
© 2010 Bennett, McRobb and Farmer1 Use Case Description Supplementary material to support Bennett, McRobb and Farmer: Object Oriented Systems Analysis.
Exploring Word Grauer and Barber 1 Committed to Shaping the Next Generation of IT Experts. Chapter 4: Advanced Features: Tables, Styles, and Sections.
INTERNATIONAL CONFERENCE ON NATURAL LANGUAGE PROCESSING NLP-AI IIIT-Hyderabad CIIL, Mysore ICON DECEMBER, 2003.
Chapter 20: Natural Language Generation Presented by: Anastasia Gorbunova LING538: Computational Linguistics, Fall 2006 Speech and Language Processing.
What is NLG? NLG "is the process of deliberately constructing a natural language text in order to meet specified communicative goals". [McDonald 1992]
Natural Language Generation Research Presentation Presenter Shamima Mithun.
Generation Miriam Butt January The Two Sides of Generation 1) Natural Language Generation (NLG) Systems which take information from some database.
Searching a Database Creating a Form for Users to Search a Database.
Natural Language Generation Martin Hassel KTH CSC Royal Institute of Technology Stockholm
A Simple Guide to Using SPSS© for Windows
With Microsoft ® Office 2010© 2011 Pearson Education, Inc. Publishing as Prentice Hall1 GO! with Microsoft ® Office 2010 Common Features Using the Common.
Natural Language Generation Ling 571 Fei Xia Week 8: 11/17/05.
Customizing Outlook. Forms Window in which you enter and view information in Outlook Outlook Form Designer The environment in which you create and customize.
Microsoft Office XP Illustrated Introductory, Enhanced Office Applications with Internet Explorer Integrating.
Data Management Seminar, 8-11th July 2008, Hamburg 1 WinW3S - Translation of Forms and Labels (Mail Merge)
Microsoft Office Word 2013 Expert Microsoft Office Word 2013 Expert Courseware # 3251 Lesson 4: Working with Forms.
Data Management Seminar, 8-11th July 2008, Hamburg WinW3S - Translation of Forms and Labels (PDF)
Laboratory Exercise # 3 – Basic File Management Office Productivity Tools 1 Laboratory Exercise # 3 Basic File Management Objectives: At the end of the.
GSE M&M WEEK 11.
1 ADVANCED MICROSOFT WORD Lesson 15 – Creating Forms and Working with Web Documents Microsoft Office 2003: Advanced.
Lecture 1, 7/21/2005Natural Language Processing1 CS60057 Speech &Natural Language Processing Autumn 2005 Lecture 1 21 July 2005.
Microsoft Expression Web - Illustrated Unit B: Creating a Web Site.
Introduction to Natural Language Generation
XP New Perspectives on Introducing Microsoft Office XP Tutorial 1 1 Introducing Microsoft Office XP Tutorial 1.
I hereby declare that this document is based on my project experience. To the best of my knowledge, this document does not contain any material that infringes.
MAIL MERGE Designing Documents with. Terms Mail Merge: A process that inserts variable information into a standardized document to produce a personalized.
Chapter 3 PART 2 - SPREADSHEET CMPF 112 : COMPUTING SKILLS CALC FOR LINUX.
Nancy Severe-Barnett Program Coordinator, SCIS
Introduction to Computational Linguistics Programming I.
Marcel Casado NCAR/RAP WEATHER WARNING TOOL NCAR.
1 Integrating Microsoft Office 2003 Tutorial 2 – Integrating Word, Excel, and Access.
Organizing Your Information
McGraw-Hill/Irwin The Interactive Computing Series © 2002 The McGraw-Hill Companies, Inc. All rights reserved. Microsoft Excel 2002 Lesson 1 Introduction.
With Microsoft ® Office e© 2013 Pearson Education, Inc. Publishing as Prentice Hall1 Common Features Using the Common Features of Microsoft ® Office.
Understanding Natural Language
Lecture Set 2 Part A: Creating an Application with Visual Studio – Solutions, Projects, Files.
A guide to creating a power point display Essentials Ctl M =New Slide: a new slide can be inserted. It is placed after the slide that you are viewing.
1 Introduction  Extensible Markup Language (XML) –Uses tags to describe the structure of a document –Simplifies the process of sharing information –Extensible.
Using Microsoft Word & Excel §Open a new document in MS Word §Click on Tools & then Mail Merge l a dialogue box, like the one on this page, will appear.
1 CS Tutorial 5 Frid. Oct 23, 2009 Design Document Tutorial.
Microsoft ® Office Excel 2003 Training Using XML in Excel SynAppSys Educational Services presents:
XP New Perspectives on Microsoft Office FrontPage 2003 Tutorial 7 1 Microsoft Office FrontPage 2003 Tutorial 7 – Creating and Using Templates in a Web.
Introduction to Computational Linguistics
Chap#11 What is User Support?
Mail Merge.
Tool for Ontology Paraphrasing, Querying and Visualization on the Semantic Web Project By Senthil Kumar K III MCA (SS)‏
SPSS- Tutorial The following power-point slides show you how to use some of the features in SPSS. A survey of 20 randomly selected companies asked them.
Jan 2004CSA3050: NLG21 CSA3050: Natural Language Generation 2 Surface Realisation Systemic Grammar Functional Unification Grammar see J&M Chapter 20.3.
1 After completing this lesson, you will be able to: Get around the Internet with your browser. Connect to the Internet. Print Web pages. Save Web pages.
Natural Language Generation Martin Hassel KTH NADA Royal Institute of Technology Stockholm
NLP. Introduction to NLP (U)nderstanding and (G)eneration Language Computer (U) Language (G)
Family Classroom Museum Suzanne Hutchins Lonna Sanderson.
Lecture Set 2 Part A: Creating an Application with Visual Studio – Solutions, Projects, Files 8/10/ :35 PM.
COMPUTER PROGRAMMING I 3.01 Apply Controls Associated With Visual Studio Form.
CSC 230 (Blum)1 Visual Basic 2005 Hello World Fall 2005 T. Blum.
© 2010 Delmar, Cengage Learning Chapter 11 Creating and Using Templates.
Certificate in Accounting NOS 116, 118, 213, 222 Lecture 1: Introduction.
NATURAL LANGUAGE PROCESSING
Discourse & Natural Language Generation Martin Hassel KTH NADA Royal Institute of Technology Stockholm
Chapter 9 The Communicative Approach.
Personal Planning System The Merging a New Plan Enterprise Edition.
WHAT ARE THE STEPS TO CREATE FOLDERS TO ORGANIZE MESSAGES IN YAHOO MAIL?
European Computer Driving Licence
Correct Order Correct Order is a game we like to use to reinforce the correct sequence or flow of a process. Use this game when a process or sequence being.
Using the Correct Order Template
6.1 Creating Charts  Data Series and Chart
European Computer Driving Licence
Presentation transcript:

Natural Language Generation: Discourse Planning Paul Gallagher 28 April 2005 (Material adapted from Ch. 20 of Jurafsky and Martin unless otherwise noted)

Introduction Programs that generate natural language are very common “Hello, world” (Kernighan and Ritchie) Template filling and mail merge Simple, but inflexible!

Introduction What do we really mean when we say “Natural Language Generation”? “the process of constructing natural language outputs from non-linguistic inputs” (Jurafsky and Martin, p. 765) Map from meaning to text (the inverse of natural language understanding)

Introduction Contrast with NLU Characteristics of an NLG system In NLU, focus is hypothesis management In NLG, focus is choice Characteristics of an NLG system Produce an appropriate range of forms Choose among forms based on internal meaning and context

Introduction What kinds of choices? Content selection Lexical selection Sentence structure Aggregation Referring expressions Discourse structure

Introduction NLG Examples Generate textual weather forecasts from weather maps Summarize statistical data from database or spreadsheet Explain medical information Authoring aids (Reiter and Dale)

Linguistic Realization Knowledge Base + Communicative Goal Meaning An NLG Reference Architecture Discourse Planner Text Plan Map Sentence Planning Sentence Plans Linguistic Realization Natural Language Text Text Adapted from Dale & Reiter. “Building Applied Natural Language Generation Systems”

Discourse Planner “Discourse planning…[imposes] ordering and structure over the set of messages to be conveyed.” (Reiter and Dale) Push or Pull? The planner selects or receives its content from the knowledge base. (McDonald) Outputs a tree structure defining order and rhetorical structure. (Reiter and Dale)

Text Schemata Observation: Many texts follows consistent structural patterns Example: Instructions For each step: Mention preconditions Describe the step Describe sub-steps Mention side-effects

Text Schemata Knowledge base representation of a saving procedure (Jurafsky and Martin. Fig. 20.5)

Text Schemata A schema from representing procedures. Implemented as an augmented transition network (ATN). Jurafsky and Martin. Fig 20.6

Text Schemata Sample output of the example: Save the document: First, choose the save option from the file menu. This causes the system to display the Save-As dialog box. Next choose the destination folder and type the filename. Finally, press the save button. This causes the system to save the document.

Rhetorical Relations Text schemata still not very flexible Schema is essentially a hard-coded text plan. There is an underlying structure to language which we can take advantage of to develop richer expressions: Rhetorical Structure Theory

Rhetorical Relations I love to collect classic automobiles. My favorite car is my 1899 Duryea. However, I prefer to drive my 1999 Toyota. nucleus Elaboration satellite Contrast satellite

Rhetorical Relations How do we apply RST to a discourse planner? Post a high-level goal to the planner (e.g., “Make the hearer competent to save a document”) Create plan operators which expand goals into sub-goals, creating a rhetorical structure tree.

Rhetorical Relations Name: Expand Purpose Effect: (COMPETENT hearer (DO-ACTION ?action)) Constraints: (AND (c-get-all-substeps ?action ?sub-actions) (NOT singular-list? ?sub-actions)) Nucleus: (COMPETENT hearer (DO-SEQUENCE ?sub-actions)) Satellites: (((RST-PURPOSE (INFORM s hearer (DO ?action))) *required*)) Name: Expand Sub-Actions Effect: (COMPETENT hearer (DO-SEQUENCE ?actions)) Constraints: NIL Nucleus: (foreach ?actions (RST-SEQUENCE (COMPETENT hearer (DO-ACTION ?actions)))) Satellites: Jurafsky and Martin, pp. 786 and 788

Rhetorical Relations The full rhetorical structure for the example text. Jurafsky and Martin. Fig. 20.7.

References Jurafsky, D. & Martin, J. H. (2000). Speech and and Language Processing. Reiter, E. and Dale, R. (1997). “Building Applied Natural Language Systems. McDonald, D. D. “Natural Language Generation”. (Appeared in Handbook of Natural Language Generation, edited by Dale, R., Moisl, H., and Somers, H.)