User:ChristianPietsch/List of NLG systems

From ACL Wiki
< User:ChristianPietsch
Revision as of 06:18, 7 December 2009 by ChristianPietsch (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

This is an attempt to represent John Bateman and Michael Zock's list of NLG systems in a wiki format.

Some works remains to be done:

Final update: This data collection is now maintained in a wiki with Semantic Mediawiki and BibWiki extensions at http://www.nlg-wiki.org/systems/

The table below remains here as an unmaintained archive only.


Table of NLG systems

BibTeX cite keys main workers starting year ending year languages generated name of system short description long description entry updated theoretical framework URL domain or application name of forerunner system unique permanent name
Kohl90, Kohl89 Kohl, Gardent, Plainfossé, Reape 1989 1990 English Accord Natural language and graphic interface Natural language and graphic interface 2000-06-03 Accord
Elhadad:1992-phd Elhadad 1992 1992 English Advisor Generation of arguments Generation of arguments 2000-06-20 arguments Advisor
Kruijff-etal-2000-coling, Bateman-etal2000-lrec, Agile-note97 Krujiff, Korbayová, Teich, Hartley, Bateman, Sharoff, Scott, Staykova, Sokolova 1999 2001 Bulgarian, Czech, Russian AGILE generates instructions in the domain of Computer-Aided Design AGILE is a tool which allows a technical author to specify, in a non-linguistic representation, the 'content' of different tasks that can be performed by users of CAD-CAM software. The AGILE system can then automatically express these content specifications in styles appropriate to different sections of a CAD-CAM manual (procedures, ready reference ...) in Bulgarian, Czech and Russian. The generated texts are displayed in a browser as hyperlinked documents. No expertise in knowledge representation is required, although some training with the interface is needed. The system has been evaluated and the results are described in the relevant project deliverables. 2001-09-18 SFG http://www.itri.bton.ac.uk/projects/agile CAD, instructions KPML AGILE
Coch-etal95, Coch96a, CochDavid94-anlp Coch, David 1993 1995 French AlethGen/La Redoute generation of one-page reply letters generation of one-page reply letters 2000-06-03 business letters AlethGen/La Redoute
Stock91 Stock 1990 1991 Italian ALFRESCO multimedia NL based information access multimedia NL based information access 2000-06-03 ALFRESCO
Kukich83 Kukich 1983 1983 English ANA generation of stock market reports generation of stock market reports 2000-01-01 stock market ANA
Blocher92 Blocher 1992 1992 German ANTLIMA-1 generating scene descriptions generating scene descriptions 2000-06-03 scene description ANTLIMA-1
deCarolis-etal98 de Carolis, de Rosis 1996 1999 Italian ARIANNA dynamic hypermedia dynamic hypermedia 2000-06-03 ARIANNA
Kim-etal02-artequakt Kim, Alani, Hall, Lewis, Millard, Shadbolt, Weal 2002 ongoing English Artequakt Automatic generation of narrative biographies of artists The Artequakt project is working towards automatically generating narrative biographies of artists from knowledge that has been extracted from the Web and maintained in a knowledge base. An overview of the system architecture is presented here and the three key components of that architecture are explained in detail, namely knowledge extraction, information management and biography construction. It is our belief that by building a story-schema layer on top of an ontology we can create dynamic stories within a certain domain. 2002-10-12 biographies Artequakt
Dalianis99 Dalianis 1996 1999 English Astrogen Generation from formal specifications and STEP/EXPRESS specifications. ASTROGEN (Aggregated deep and Surface naTuRal language GENerator) is a Natural Language Generator written in Prolog. Which hopefully can be used by almost anybody. ASTROGEN has been used for generation of natural language (English) from formal specifications and STEP/EXPRESS Specifications. ASTROGEN consists basically of two modules the Deep and the Surface generator. ASTROGEN was written originally for generating Natural Language from formal specifications for the telecom domain. <img src=../images/astrogen.gif > 2000-06-03 http://www.dsv.su.se/~hercules/ASTROGEN/ASTROGEN.html telecom Astrogen
Vavargard2000 Vargagard 1999 1999 English, Swedish Autotext Weather forecast report generator Weather forecast report generator 2000-06-20 weather Autotext
Goldman75 Goldman 1975 1975 English BABEL paraphrase generation & lexical choice paraphrase generation & lexical choice on the basis of abstract representations (conceptual dependencies) 2000-01-01 BABEL
Kronfeld:1986 Kronfeld 1986 1986 English BERTRAND generation of definite descriptions generation of definite descriptions 2000-01-01 BERTRAND
Weiner80 Weiner 1980 1980 English BLAH generation of explanations concerning income taxes generation of explanations concerning income taxes 2000-01-01 income tax BLAH
Mittal-etal98, Mittal-etal95 Mittal, Moore, Carenini, Roth, Mattis 1995 1998 English Caption Generation System generates captions to complex automatically produced graphical presentations generates captions to complex automatically produced graphical presentations 2000-06-03 FUF figure captions FUF Caption Generation System
sharp-88 Sharp 1990 1995 EU CAT-2 formalism used for MT, includes analysis and generation capabilities fro the Eurotra Intermediate Structures. formalism used for MT, includes analysis and generation capabilities fro the Eurotra Intermediate Structures. 2000-06-03 http://www.wh2.tu-dresden.de/~dagmar/cat2/ CAT-2
Jokinen94 Jokinen 1994 1994 English CDM generation of explanatory dialogues generation of explanatory dialogues 2000-01-01 explanation CDM
Kan-etal01, Kan-etal01-doc Kan, McKeown, Klavans 2000 ongoing English CENTRIFUSER A multidocument summarization system Centrifuser is a multidocument summarization system for documents which are in the same domain and genre. Centrifuser's indicative module employs NLG techniques to produce text that discusses groups of documents and their metadata properties by mimicking library card catalog entries. 2001-11-02 http://centrifuser.cs.columbia.edu/ summaries CENTRIFUSER
Kohl, Eisele, Momma, Dörre, Geltz, Netter, Kärcher, Tappe, Frank, Meier, van Geenhoven 1989 1992 French, German CHARON grammar development (allegedly including generation) for LFG grammar development (allegedly including generation) for LFG 2000-06-03 LFG http://www.dfki.de/lt/registry/parsers/charon.html CHARON
Freedman98 Freedman 1998 1998 English Circsum-Tutor tutoring system tutoring system 2000-06-03 tutoring Circsum-Tutor
AlshawiPulman92 Alshawi, Pulman 1992 1992 English, Swedish, French, Spanish CLE Core Language Engineering: mostly analysis oriented with some generation work Core Language Engineering: mostly analysis oriented with some generation work 2000-06-03 CLE
Evans-etal2006, Piwek-etal98, Piwek-etal99-itri, Piwek-etal99 Piwek, Evans, Power 1998 2001 English, French CLIME Automated responses to queries, posed in natural language, of any prescriptive semi-formal documentation CLIME is a system to provide automated responses to queries, posed in natural language, of any semi-formal documentation of a prescriptive nature, such as codes of practice and other quasi-legal documentation. The CLIME project takes advantage of the semi-regular nature of this information to allow the integration of working NL software with existing expertise in codifying legal reasoning, to provide a usable and useful system. 2001-09-18 http://www.bmtech.co.uk/clime documentation CLIME
Gedalia, Elhadad 1999 2000 English CLINT hybrid template/word-based generator hybrid template/word-based generator 2000-06-03 http://www.cs.bgu.ac.il/~elhadad/clint.html CLINT
CaldwellWhite97, WhiteCaldwell97-anlp, WhiteCaldwell98 White, Caldwell 1997 1998 English CogentHelp on-line help for GUIs Semi-automatic generation of on-line help for GUIs, with authoring interface for human-written "snippets". Additional information at: <A href=http://www.cogentex.com/research/cogenthelp/index.shtml>http://www.cogentex.com/research/cogenthelp/index.shtml</A> 2000-09-19 http://www.cogentex.com/systems/cogenthelp/ online help CogentHelp
vandeRiet-etal98 van de Riet, Burg, Dehne 1995 1998 English Color-X NLG for verification of models of information and communication systems In Color-X, tools were developed for building executable prototypes of information and communication systems from specifications expressed in natural language. Papers referenced here refer to the addition of NLG capabilities for verification of the models and requirements. 2000-06-20 FG communication systems Color-X
McKeown-etal90 McKeown et al. 1990 1990 English COMET generation of explanations (multimodal output) generation of explanations (multimodal output) 2000-01-01 FUF explanation FUF COMET
Sigurd83, Sigurd84 Sigurd 1983 1984 Swedish COMMENTATOR generation of scene descriptions generation of scene descriptions 2000-01-01 scene description COMMENTATOR
Fawcett:1988, Fawcett90-nlgw5 Fawcett, Tucker 1988 1992 English COMMUNAL generation with systemic functional grammar generation with systemic functional grammar 2000-01-01 SFG COMMUNAL
Geldof99-ki Geldof 1999 1999 English COMRIS meeting assistant meeting assistant 2000-10-01 appointments COMRIS
Marsi01, Marsi98, Marsi95, Marsi98-text Marsi 1995 2000 Dutch CONPAS Content-to-Prosodically Adequate Speech CONPAS (Content-to-Prosodically Adequate Speech) is a front-end generator for producing spoken output with appropriate prosody. It consists of three modules: SynR, a module for syntactic realization, ProsR, a module for prosodic realization and, SpeechS, a module for speech synthesis. It is driven by a planning component (PlanR) and produces spoken output in a number of domains; the text planner is specific for each domain. Particular emphasis was placed on the pitch accent placement, and this has been evaluated. 2001-09-17 prosody generation CONPAS
Smadja:1991, SmadjaMcKeown90 Smadja, McKeown 1990 1992 English COOK use of collocation for generating stock market reports use of collocation for generating stock market reports 2000-01-01 stock market COOK
McKeown:1979 McKeown 1979 1979 English Co-Op Cooperative answer generation Cooperative answer generation 2000-06-20 Co-Op
LevineMellish94 Mellish, Levine 1994 1998 English CORECT Generate documents for requirements definition in the design of electronic test equipment a system using NLG techniques to generate documents for requirements definition in the design of electronic test equipment; developed jointly with Racal Research, Racal Instruments, the Unis of Edinburgh and Sussex, and Intelligent Applications Ltd. Builds on its predesessor, the IDAS system. 2000-06-03 http://www.cogs.susx.ac.uk/lab/hct/corect/ electronics IDAS CORECT
Bus:Dec:Dia:97 Busemann, Declerck, Kader, Diagne, Dini, Klein, Schmeier 1997 1997 German Cosma Appointment scheduling agent Appointment scheduling agent 2000-06-03 appointments TG/2 Cosma
NealShapiro88 Neal, Shapiro 1988 1988 English Cubricon Generation in multimedia interfaces Generation in multimedia interfaces 2000-06-20 Cubricon
Theune-etal01-jle Theune, Klabbers, Odijk, de Pijper, Krahmer 1998 ongoing independent D2S System for producing spoken language from data D2S is a system for the creation of data-to-speech systems in different languages and domains. The most important characteristic of a data-to-speech system is that it combines language and speech generation: language generation is used to produce a natural language text expressing the system's input data, and speech generation is used to make this text audible. In D2S, this combination is exploited by using linguistic information available in the language generation module for the computation of prosody. This allows us to achieve a better prosodic output quality than can be achieved in a plain text-to-speech system. For language generation in D2S, the use of syntactically enriched templates is guided by knowledge of the discourse context, while for speech generation pre-recorded phrases are combined in a prosodically sophisticated manner. This combination of techniques makes it possible to create linguistically sound but efficient systems with high quality language and speech output. 2001-09-30 spoken dialogue LGM D2S
Bateman-etal01-dart, Bateman-etal98-inlgw, Reichenberger-etal95-acm, Reichenberger-etal95-ave Bateman, Alexa, Teich, Kamps, Reichenberger, Kleinz, Rondhuis 1994 1996 English Dart-bio Automatic generation of pages combining textual artist bibliographies and timeline graphs concerning the Bauhaus Automatic generation of pages combining textual artist bibliographies and timeline graphs concerning the Bauhaus. Producing formatted pages in response to selections of artists made by the user with a graphical information retrieval interface.</p><p>Generation was performed using the KOMET generation system. An example page generated from a very large knowledge base compiled from the Dictionary of Art at that time is:</p><p><img src=../images/pic1-layout.gif width=700 height=525></p> 2000-06-03 SFG biographies, art history, layout and text KOMET Dart-bio
Mueller90 Mueller 1983 1988 English Daydreamer generator A recursive-descent generator for reporting stream of thought A recursive-descent generator for reporting stream of thought. The source code of the program can be found <a href=http://www.panix.com/~erik/programs/daydreamer/src/dd_gen.cl>here</a>. 2001-12-05 http://www.panix.com/~erik/programs/daydreamer/dd.htm Daydreamer generator
Sgouros99 Bailey, Mellish, Ritchie, Sgouros 1997 1999 English DEFACTO DEFACT (DEsign Framework for interACTive stOry systems) an interactive system for generating stories. DEFACT (DEsign Framework for interACTive stOry systems) an interactive system for generating stories. It seeks to develop and test a prototype system which incorporates Artificial Intelligence techniques into interactive, multi-media story generation and presentation, in order to create a richer, and therefore more satisfying, experience for its user/s. 2000-06-03 http://www.dai.ed.ac.uk/daidb/infweb/people/homes/paulba/defacto/ stories DEFACTO
Sprenger93, Horacek93 Müller, Sprenger, Horacek 1990 1994 German DIAMOD KADS-based expert system explanation KADS-based expert system explanation 2000-06-03 explanation DIAMOD
WendholtNovak93 Novak, Wendholt 1993 1993 German DING incremental generation incremental generation 2000-06-03 DING
NirenburgNirenburg88-coling, Nirenburg-etal89 Nirenburg 1987 1988 English DIOGENES generation for machine-translation generation for machine-translation 2000-01-01 DIOGENES
Uszkoreit-etal94, Bus:Dec:Dia:97 Uszkoreit, Busemann 1994 1998 German DISCO An appointment scheduling agent using natural language An appointment scheduling agent using natural language 2000-06-20 appointments DISCO
Seligman91-phd Seligman 1985 1991 English Discourser Structured discourse System developed in a dissertation that presents two components. First: An inheritance-based grammar which represents syntactic and semantic structures as well as the mapping relations between these as frames. The approach to sentence generation features the merging of structures via inheritance and the tracking of expressive choices using 'multiple worlds'. Second: new semantic/pragmatic representations for paragraph-length discourses, and expression procedures based upon them. Discourse relations are discovered via judgements of synonomy, and are arranged in lattices (rather than exclusively in hierarchies). A small set of network traversal strategies permits the repeated application of the sentence generator to generate paragraphs expressing the whole net. A facility for expressing 'situational formulas' adds a pragmatic component to the system. 2001-09-26 inheritance grammar Discourser
Gruber-etal95 Gruber, Vemuri, Rice 1992 1996 English DME-HTW A demonstration of 'virtual documents', generated on demand in response to the user's selections A demonstration of 'virtual documents', generated on demand in response to the user's selections 2000-06-03 http://www-ksl.stanford.edu/people/gruber/virtual-documents-htw/top-level.html DME-HTW
HoughtonPearson88 Houghton, Pearson 1988 1988 English DORIS planning dialogues (vocal output) planning dialogues (vocal output) 2000-01-01 DORIS
Drafter95 Scott, Hartley, Power, Paris, Vander Linden, Evans 1995 1995 English, French Drafter-I helps technical authors write instruction documents helps technical authors write instruction documents 2000-06-03 SFG instructions KPML Drafter-I
ScottEvans98, Power-etal98 Scott, Evans, Power 1998 2000 English, French, Italian Drafter-II multilingual generation of instructions multilingual generation of instructions 2000-06-03 instructions Drafter-II
Deemter:Odijk:97, vanDeemter97 van Deemter, Leermakers, Odijk 1994 1997 English DYD generates a spoken monologue describing a selected music composition and its recording DYD (Dial Your Disc) is an interactive system that supports browsing through a large music database (Mozart compositions); each time a track is selected, it generates a spoken monologue describing the composition and its recording 2000-06-03 music DYD
Geldof98, GeldofVelde97-ewnlg, GeldofVelde97-aaai, GeldofVandeVelde97 Geldof, van de Velde 1997 1997 English Ecran generates descriptions of movies, using a model of the user's interests generates descriptions of movies, using a model of the user's interests 2000-06-03 templates http://arti.vub.ac.be/AAAI97/ecran.html movie description Ecran
Cawsey90 Cawsey 1990 1990 English EDGE plan-based generation of explanatory discourse plan-based generation of explanatory discourse 2000-01-01 http://www.cee.hw.ac.uk/~alison/AAAI.ps explanation EDGE
Claassen92 Claassen 1992 1992 English Edward Refering expressions in a multimedia environment Refering expressions in a multimedia environment 2000-06-20 Edward
MooreIJCAI89, EES-87 Moore, Swartout, Paris 1986 1990 English EES generation of explanations generation of explanations 2000-01-01 SFG explanation Penman EES
Poller-etal96, PollerHeisterkamp96 Poller 1993 1996 German, English EFFENDI Efficient Formulation of Dialogue Contributions for train inquiries Efficient Formulation of Dialogue Contributions: development and implementation of an efficient How-to-Say component for the generation of system utterances of a speech dialogue system for train inquiries. Extends the TAG-GEN generator developed within the WIP system 2000-06-03 TAG train inquiries TAG-GEN EFFENDI
Weizenbaum66 Weizenbaum 1966 1966 English ELIZA simulation of a dialog with a psychotherapist simulation of a dialog with a psychotherapist 2000-01-01 psychiatry ELIZA
<i>unclaimed</i> ELU Patr-II based MT, analysis and generation Patr-II based MT, analysis and generation 2000-06-03 ELU
Dale90-dale, Dale89 Dale 1988 1990 English EPICURE generation of cooking recipes generation of cooking recipes 2000-01-01 recipes EPICURE
Clippinger Clippinger 1977 1977 English ERMA simulation of performance errors (dialog with a psychiatrist) simulation of performance errors (dialog with a psychiatrist) 2000-01-01 psychiatry ERMA
CaldwellKorelsky94 Caldwell, Korelsky 1994 1994 English EXCLASS generation of job descriptions generation of job descriptions 2000-01-01 MTM job descriptions EXCLASS
WhiteCaldwell98 White, Caldwell 1998 1998 English Exemplars dynamic HTML page generation on the basis of a classification hierarchy for text planning dynamic HTML page generation on the basis of a classification hierarchy for text planning 2000-06-03 Exemplars
Chester76 Chester 1976 1976 English Expound translates formal poofs into English translates formal poofs into English 2000-06-20 formal proofs Expound
BangaloreRambow2000, Bangalore-etal2000-learning, Bangalore-etal2000-evaluation, BangaloreRambow2000-corpus Bangalore, Rambow 2000 2002 English FERGUS Flexible Empiricist/Rationalist Generation Using Syntax Extends statistically guided generation using n-grams with a tree-based statistical model as well as a traditional tree-based syntactic grammar based on the wide coverage XTAG grammar 2002-11-17 statistical independent FERGUS
Ward94, Ward92 Ward 1992 1992 English FIG connectionist generator connectionist generator 2000-01-01 http://www.sanpo.t.u-tokyo.ac.jp/~nigel/clg.html FIG
MeunierDanlos98 Meunier, Danlos 1998 1998 French, English Flaubert multilingual generator multilingual generator 2000-06-03 TAG Flaubert
Reiter90-dale, MarksReiter90, Reiter90, Reiter91-ci Reiter 1990 1990 English FN avoiding false implicatures avoiding false implicatures 2001-09-05 FN
Goldberg-etal94, Bourbeau-etal90-coling, Goldberg-etal94-ieee Goldberg, Kittredge, Polguère, Driedger 1989 2000 English, French FoG FoG (Forecast Generator) provides bilingual generation of operational weather forecast reports. FoG (Forecast Generator) provides bilingual generation of operational weather forecast reports. The first operational forecast produced by FoG was issued on Feb.27, 1992. 2000-01-01 MTM weather FoG
EndrissKlabunde2000 Endriss, Klabunde 2000 2000 German FOGS generates narratives with a focus background structure The system (acronym for "focus generation system") generates renarratives with a focus background structure (more specifically, the single sentences are annotated with prosodic features to indicate which element functions as bearer of the respective focal accent). 2000-06-20 stories FOGS
Contant86, Contant:1985 Contant 1985 1986 French FRANA French version of ANA French version of ANA 2000-01-01 stock market ANA FRANA
Emele86, Emele87-gwai, HeidMomma87 Emele, Heid, Momma 1986 1987 German FREGE object-oriented front-end generator object-oriented front-end generator 2000-06-03 FREGE
ElhadadRobin92, Elhadad93-fuf, ElhadadRobin96 Elhadad 1990 ongoing independent, English, Spanish, Hebrew FUF implementation of Functional Unification Grammar implementation of Functional Unification Grammar 2000-01-01 FUF http://www.cs.bgu.ac.il/research/projects/surge/index.htm independent FUF
Peyragrosse:1993 Peyragrosse 1993 1993  ?? GASPAR Paraphrase of SQL queries Paraphrase of SQL queries 2002-09-20 DB queries GASPAR
Etchegoyhen98 Etchegoyhen 1998 1998 English GBgen GB-based generator GB-based generator 2000-06-03 GB GBgen
Adachi97 Adachi 1997 1997 GCD cooking definitions cooking definitions 2000-06-03 recipes GCD
Mitkov90-geo Mitkov 1990 1990 Bulgarian GECO production of geometrical descriptions production of geometrical descriptions 2000-01-01 geometry GECO
Henschel-etal02-konvens, Henschel02-gem-manual Delin, Bateman, Henschel, Allen 1998 2002 English GeM multimodal generation GeM (Genre and Multimodality) investigated the relationship between multimodal document genres and their realization in text, layout and diagrams. As part of this work a prototype page generation system was developed based on XML, XSLT and the GeM project multimodal annotation scheme. The prototype concentrates specifically on the controlled generation of the layout (including placing of text and graphics, titles, and font selections) adopted on the page, making this sensitive to generic constraints holding for the multimodal document type; the generation of the language elements included is not in focus. The generic constraints are derived from the corpus analysis done within the GeM project and are represented following the GeM annotation scheme. <P>Two examples of generated pages from the prototype are the following, the first from the genre of bird field guides and the second from the genre of instruction manuals.<P> <table><tr><td><img src=../images/gem-bird-konvens.gif ></td><td><img src=../images/gem-pegasys-konvens.gif > </td></tr></table> 2002-09-28 SFG http://purl.org/net/gem layout and text GeM
Conklin:1987 Conklin 1983 1987 English GENARO description of pictures description of pictures 2000-01-01 TAG picture description GENARO
Glass-etal94 Glass, Seneff, Polifroni 1994 1994 English GENESIS multilingual spoken instructions multilingual spoken instructions 2000-06-03 instructions GENESIS
Fawcett90-nlgw5, FawcettTucker90-coling Fawcett, Tucker 1990 1994 English GENESYS the large systemic functional grammar component of the Communal system the large systemic functional grammar component of the Communal system 2000-01-01 SFG GENESYS
Shieber88-coling Shieber 1987 1988 English GeneSys Experimental generation Experimental generation 2002-05-02 GeneSys
deCarolis-etal96, deCarolis-etal97 de Carolis, de Rosis, Pizzutilo, Grasso, Rossiello, Berry, Gillie 1996 1997 Italian GeNet dynamic hypermedia generation the medical domain and tailored to the user dynamic hypermedia generation the medical domain and tailored to the user 2000-06-03 medical GeNet
Teunissen97 Teunissen 1997 1997 English GENIUSS generator based on universal semantic syntax generator based on universal semantic syntax 2000-06-03 GENIUSS
TomitaNyberg88 Tomita, Nyberg 1988 1988 English? GenKit 2001-06-01 GenKit
Maybury89 Maybury 1989 1989 English GENNY use of RST relations for producing coherent texts use of RST relations for producing coherent texts 2000-01-01 GENNY
Kehl86 Kehl 1986 1986 German GEOTEX description of geometrical configurations description of geometrical configurations 2000-01-01 geometry GEOTEX
Kehl89 Kehl 1989 1989 English, German GEOTEX-E bilingual version of Geotex bilingual version of Geotex 2000-06-03 geometry GEOTEX GEOTEX-E
Pemberton89 Pemberton 1989 1989 English GESTER story generation story generation 2000-01-01 stories GESTER
Marchant-etal96 Marchant, Mellish, Cerbah, Levine 1995 1996 English GhostWriter generation assisting human technical authors in the production of aircraft manuals a demonstration of how plan-based modelling and NLG can be used to assist human technical authors in the production of aircraft manuals 2000-06-03 http://www.aiai.ed.ac.uk/~johnl/ghostwriter/ manuals GhostWriter
Williams2004-nlg, Williams02-connectives, WilliamsReiter03 Williams 2000 2003 English GIRL Microplanning discourse-level choices for people with poor literacy System for generating appropriate texts for people with different reading levels 2006-03-09 http://www.csd.abdn.ac.uk/~swilliam/phd.html Personalised basic skills summary reports GIRL
gistgenerator Swartout 1983 1983 English GIST explanation of the results of a medical expert system explanation of the results of a medical expert system 2000-01-01 medical GIST-medical
NotStock94, ParisScott94-inlgw, PowerCavallotto96-gist, PianesiPianta94-gist, VanderLindenScott95-ijcai, Not96-mercatali, Pemberton-etal94, HartleyParis97-MT, Bateman-etal94-ecai, Delin-etal96, MatiasekTrost96, MatiasekBuchberger95 Cavallotto, Magnini, Lavid, Matiasek, Not, Paris, Pianesi, Pianta, Power, Scott, Stock, Trost, Vander Linden 1994 1996 English, German, Italian GIST generates instructions for filling in tax forms The GIST project addresses the construction of a multilingual generation system for texts describing administrative procedures (e.g., how to apply for a social security facility, what to do in order to get visas, etc.) starting from language independent specifications. We considered three languages: English, German and Italian. The application domain of the developed prototype is represented by instructions on how to fill out pension forms. <P>A longer set of references is available <A HREF=http://ecate.itc.it:1025/projects/gist/gist-bibliography.html> here</A>. 2000-09-18 SFG, FUF, GB http://ecate.itc.it:1025/projects/gist.html tax forms GIST-tax
Kantrowitz92 Kantrowitz, Bates 1992 1992 English GLINDA narration and intercharacter communication narration and intercharacter communication 2000-01-01 GLINDA
Klabbers-etal98, TheuneKlabbers98-inlg Klabbers, Odijk, de Pijper, Theune 1996 1995 Dutch GoalGetter GoalGetter generates spoken summaries of football (soccer) matches on the basis of tabular data GoalGetter generates spoken summaries of football (soccer) matches on the basis of tabular data 2000-06-03 http://iris19.ipo.tue.nl:9000/index.html football GoalGetter
Iordanskaja-etal88-coling Iordanskaja et al. 1988 1988 English GOSSIP implementation of the meaning-text theory, praphrase An implementation of the meaning-text theory, producing paraphrase 2000-01-01 GOSSIP
Rich79 Rich 1979 1979 English GRUNDY user modelling via stereotypes user modelling via stereotypes 2000-06-20 user modelling GRUNDY
Langkilde:2000 Langkilde 2000 2000 English Halogen Statistical generation from interlinguas using a packed forest intermediate representation Statistical generation from interlinguas using a packed forest intermediate representation 2000-06-03 http://www.isi.edu/natural-language/projects/nlg-publications.html Halogen
wahlster-ijcai, Wahlster78, Hoeppner-etal83, Hoeppner84 Wahlster et al. 1978 1983 German HAM-ANS (HAMburg Application-oriented Natural-language System) (HAMburg Application-oriented Natural-language System) An interactive system where the determination of the adequate level of a response in dialogue interactions played a central role. 2000-01-01 HAM-ANS
jameson80, vonHahn-etal80, Wahlster78 Jameson, Hoeppner, von Hahn, Wahlster 1980 1980 German HAM-RPM hotel manager hotel manager 2000-06-03 reservations HAM-RPM
DiMarco-etal95, WannerHovy96, DiMarco-etal97 DiMarco, Hovy, Hirst, Wanner 1995 1998 English HealthDoc generation of medical information generation of medical information 2000-06-03 SFG medical KPML HealthDoc
Gailly88-hermes, Gailly89-phd, Gailly88-quantifier Gailly 1988 1990 French HERMES processing of quantifier scoping processing of quantifier scoping 2000-01-01 HERMES
Kaplan-etal93 Kaplan, Fenwick, Chen 1993 1993 English HIPERFLEX navigating hypertext navigating hypertext 2000-06-20 HIPERFLEX
Benelli-etal99 Benelli, Bianchi, Marti, Not, Sennati, Mellish, O'Donnell 1998 2001 English HIPS developing new interaction paradigms for navigating physical spaces The HIPS project aims at developing new interaction paradigms for navigating physical spaces. The objective of the project is to enrich the "user experience" of a city by overlapping a further dimension with the physical space: contextual and personalised information on the human environment. The project was originally designed to include NLG elements. 2001-11-03 http://marconi.ltt.dii.unisi.it/progetti/HIPS/ spatial relationships HIPS
Cross92-lexis, Cross92-thesis Cross 1994 1994 English Horace generation of water-cycle texts as an experiment in register-based generation generation of water-cycle texts as an experiment in register-based generation 2000-06-03 SFG physics Horace
NetzerElhadad98-hebrew Netzer, Elhadad 1997 1998 Hebrew HUGG A syntactic realization component for Hebrew A syntactic realization component for Hebrew, where the form of the input is intended to be as similar as possible to that for FUF/SURGE for English. The places where this does not appear possible suggest interesting places of linguistic contrast, as well as raising questions about the possible modularities involved in the generation task. 2002-10-12 FUF FUF HUGG
BontchevaWilks01-IJCAI, BontchevaWilks99, Bontcheva01-um, Bontcheva01-impact, Bontcheva02-diss Bontcheva 1997 2001 English Hylite+ Web-based system that generates adaptive hypertext explanations using a generic user modelling framework. Architecture based on a new recursive pipeline model. Web-based system that generates adaptive hypertext explanations using a generic user modelling framework. <P>The goal of our dynamic hypertext generation system, HYLITE+, is to produce encyclopaedia-style explanations of domain terminology. The corpus analysis of online encyclopaedia and previous empirical studies have shown the positive effect of additional information -- e.g., definition of key vocabulary, less-technical content, supply of background information and illustrations -- on the subjects' reading comprehension and reading behaviour. On the other hand, hypertext usability studies have shown that people read 25 per cent slower on the screen, so hypertext needs to be concise with formatting, that facilitates skimming. Our empirical studies have shown that users prefer different additional information depending on the chosen formatting and desired explanation length.<P>The majority of existing language generation systems have a pipeline architecture which offers efficient sequential execution of modules, but does not allow decisions about text content to be revised in later stages. However, in some cases choosing appropriate content can depend on text length and formatting, which in a pipeline architecture are determined after content planning is completed. Unlike pipelines, interleaved and revisionbased architectures can deal with such dependencies but tend to be more expensive computationally. Since our system needs to generate acceptable hypertext explanations reliably and quickly, the pipeline architecture was modified instead to allow additional content to be requested in later stages of the generation process if necessary. in such an architecture it is not possible to take into account text formatting and length (which are determined towards the end) when choosing the text content (which happens in the beginning).<P><H3>Architecture</H3><IMG SRC=../images/hylite-arch.gif> 2002-10-12 http://www.dcs.shef.ac.uk/~kalina Hylite+
Springer-etal91 Springer, Buta, Wolf 1991 1991 English ICG composition of business letters composition of business letters 2000-06-20 business letters ICG
McCoy-etal96-call McCoy, Pennington, Suri 1996 1996 English ICICLE teaches the syntax of written English to deaf people teaches the syntax of written English to deaf people 2000-06-03 teaching syntax ICICLE
Bouayad-Agha-etal2000-inlg, Bouayad2000-lrec, Bouayad-Agha-etal2000-idj, Bouayad-Agha99, Power2000-coling, Power2000-itri, Power-etal99-englw Power, Bouayad-Agha, Scott 1997 2000 English ICONOCLAST Combining layout and linguistic constraints In the ICONOCLAST project, we developed a framework for integrating constraints on the style and layout of the output documents in an NLG system. By interacting with the system, authors are able to determine the optimal set of constraints whose interaction will lead to the production of documents in the desired style and layout. 2000-09-18 http://www.itri.bton.ac.uk/projects/iconoclast/ layout and text, patient information leaflets ICONOCLAST
RuthPeroutka93, Peroutka-etal98 Ruth, Peroutka 1993 1998 English ICWF Interactive Computer Worded Forecast (Interactive Computer Worded Forecast) further development of the IFPS system (in which it might now be incorporated?) 2000-06-20 weather IFPS ICWF
Reiter-etal95, Reiter-etal92, ReiterMellish93 Reiter et al 1992 1992 English IDAS generation of on-line help documentation generation of on-line help documentation 2001-09-05 http://www.dai.ed.ac.uk/homes/chrism/idas.html documentation IDAS
HallerShapiro96 Haller, Shapiro 1996 1996 English IDP discourse planner discourse planner 2000-06-03 IDP
Glahn70 Glahn, Peroutka, Ruth 1970 2000 English IFPS Weather forecast report generator (template-based) Weather forecast report generator (template-based) "Interactive Forecast Preparation System". Early weather report generation still on the point of being deployed commercially. Also led on to the ICWF system (now incorporated?) .IFPS is in the template-based category, much of it written in Fortran. Does not claim an AI-basis, but must do some kind of text planning/organisation. 2000-06-20 http://www.nws.noaa.gov/tdl/icwf/ifp.htm weather IFPS
Rubinoff2000, Rubinoff92 Rubinoff 1992 1999 English IGEN Weather forecasts Weather forecasts 2000-01-01 weather IGEN
ODonnell-etal01, ODonnell-etal2000-inlg, Oberlander-etal97, Dale-etal97, Milosavljevic99-phd, Dale-etal98, MilosavljevicOberlander98, Hitzeman-etal97, Mellish-etal98-opportunism Oberlander, Mellish, O'Donnell, Knott 1995 1998 English ILEX a system for generating labels for items in an electronic museum or retail catalogue ILEX is a system for generating labels for items in an electronic museum or retail catalogue, in such a way as to reflect the interest of the user and also opportunistically to further certain educational goals. Developed in cooperation with the National Museums of Scotland, Interactive Information and VIS Interactive Media.<P>The current version of the system, Ilex2.0, produces textual descriptions of objects in a museum gallery. The user is presented with a set of objects, and chooses objects to be described by clicking, as in a conventional hypertext. When an object is selected, a description is produced on-line, which is tailored to the user and the communicative context.<P>An online demonstration is available <A href=http://www.hcrc.ed.ac.uk/ilex/demos/museum.cgi>here</A>. 2002-09-15 http://www.hcrc.ed.ac.uk/ilex museums WAG ILEX
Bates, Ingria 1981 1981 English ILIAD CALL for transformational grammar CALL for transformational grammar 2000-01-01 teaching syntax ILIAD
Vanderlinden-etal92, VanderLinden93-thesis Vander Linden 1992 1992 English IMAGENE instructions for operating telephones instructions for operating telephones 2000-01-01 SFG instructions Penman IMAGENE
Jameson87-book Jameson 1987 1987 English IMP tailoring the content by taking the user into account tailoring the content by taking the user into account 2000-01-01 user modelling IMP
Rehm99 Rehm 1998 1999 German IPaGe Incremental Parallell Generator Generates descriptions of a construction kit on the basis of conceptual structures. Generation is incremental. 2002-04-15 description IPaGe
de-Smedt90 de Smedt 1990 1990 Dutch IPF parallel-incremental sentence generation parallel-incremental sentence generation 2000-01-01 IPF
Kempen79 Kempen, Hoenkamp 1979 1979 Dutch IPG simulation of human, incremental sentence production simulation of human, incremental sentence production 2000-01-01 IPG
Colineau-etal02, Paris-etal01-anlpw, Paris-etal01-hci, Lu-etal99, Paris-etal98-adcs, Paris-etal00-impacts Paris, Vander Linden, Lu 1999 2002 English ISOLDE automatic generation of software instructions The Integrated Software and On-Line Documentation Environment (ISOLDE) is a novel, integrated software environment developed by the IIT group at CSIRO with Calvin College in the United States, and is partially funded by the US Office of Naval Research.<P>ISOLDE is providing the opportunity to streamline the software production and software documentation process, while facilitating the development of a usable, easily upgraded application. Inherent in the design is the ability to re-use software in upgrades or to totally new generation applications, with dynamically updated documentation and interaction models to match the changes. Automatic and semi-automatic task modelling forms an additional important area of work within this project. The system has been evaluated. <P><H3>Architecture</H3><IMG SRC=../images/isolde-Archit2.gif> 2000-06-03 SFG http://www.cmis.csiro.au/iit/Projects/Isolde/index.htm instructions ISOLDE
BinstedRitchie94, BinsteadRitchie96 Binsted, Ritchie 1994 1996 English JAPE JAPE (Joke Analysis and Production Engine) creates riddles using a rule-based system derived from humour theory. JAPE (Joke Analysis and Production Engine) creates riddles using a rule-based system derived from humour theory. 2000-06-03 jokes JAPE
RambowKorelsky92 Rambow, Korelsky 1992 1992 English JOYCE generates texts about software components and their security on the basis of a specifications for secure, distributed software systems generates texts about software components and their security on the basis of a specifications for secure, distributed software systems 2000-06-03 software JOYCE
Mauldin:1984 Mauldin 1984 1984 English KAFKA sentence generation from a semantic network sentence generation from a semantic network 2000-01-01 KAFKA
ClineNutter94 Cline 1994 1994 English KALOS Knowledge-based generation with revision Knowledge-based generation with revision 2000-01-01 KALOS
Appelt85-AI, Appelt88-plan Appelt 1983 1985 English KAMP modeling of the user's beliefs and goals modeling of the user's beliefs and goals 2000-01-01 KAMP
MannMoore81 Mann, Moore 1981 1981 English KDS generation of paragraphs generation of paragraphs 2000-01-01 KDS
Jacobs87-nlgw Jacobs 1987 1987 English KING help system for UNIX help system for UNIX 2000-01-01 online help KING
 ?? Kiss B.V. 1995 1999 Dutch, English, German, French KISS-Lexicalizer generation of natural language descriptions of OO-models generation of natural language descriptions of OO-models 2000-10-01 FG? DB and OO descriptions KISS-Lexicalizer
Peters93-kleist, Peters-etal91 Peters, Rutz, Siegel, Novak, Hoeppner 1991 1993 German, Japanese Kleist multilingual generation of route descriptions multilingual generation of route descriptions 2000-06-03 LFG route descriptions Kleist
LesterPorter97 Lester, Porter 1997 1997 English KNIGHT explains complex biological phenomena explains complex biological phenomena 2000-06-03 biology KNIGHT
BatemanTeich95, Alexa-etal94-sant, Alexa-etal96-ercim, Teich99-dis, Teich-etal96, Degand96-enlgw Bateman, Teich, Maier, Wanner, Henschel, Degand 1991 1996 English, German, Dutch KOMET Multilingual generation with systemic functional grammar multilingual generation with systemic functional grammar 2000-01-01 SFG KPML KOMET
Bateman97-jle, Bateman-etal99-aai, Bateman-etal91-ijcai, BatemanSharoff98 Bateman, Matthiessen, Zeng 1993 ongoing English, German, French, Dutch, Japanese, Czech, Russian, Bulgarian, Spanish, Greek, Chinese, independent KPML large-scale multilingual generation and development environment The KPML system is an ongoing development drawing on over a decade of experience in large-scale grammar development work for natural language generation (NLG). It is currently maintained at the University of Bremen, Faculty of Linguistics and Literature. The system is a graphically-based development environment for the construction, maintenance, and use of large-scale grammars written with the framework of Systemic-Functional Linguistics.<P> KPML offers a robust, mature platform for large-scale grammar engineering that is particularly oriented to multilingual grammar development and generation.Grammars have been developed using KPML for a variety of languages including English, German, Dutch, Chinese, Spanish, Russian, Bulgarian, and Czech. Many of these grammars are freely available for further research and development work within the NLG community. The English grammar is the very large Nigel grammar under development since the early 1980s in the Penman project at USC/ISI, Los Angeles. <P> A variety of configurations of KPML are available, suitable for use both by absolute beginners in a teaching environment and by large-scale grammar work. The variations are summarised on the main KPML page. KPML is written in ANSI Common Lisp, with the graphical interface in the Common Lisp Interface Manager (CLIM). The source code may be compiled and installed in full using commercial Common Lisp products supporting Common Lisp and CLIM under Unix and Windows. There are also standalone executable images available for Windows systems that may be downloaded and used without compilation and without requiring an additional Lisp software licence. 2000-06-03 SFG http://purl.org/net/kpml independent Penman KPML
Iordanskaja-etal92 Iordanskaja et al. 1992 1992 English, French LFS generates summaries of statistical data in the domain of labour market generates summaries of statistical data in the domain of labour market 2000-06-03 MTM labor market LFS
Klabbers-etal98 Klabbers, Theune, Krahmer 1997 2000 Dutch, English LGM The generation module of D2S The generation module of D2S 2001-11-02 TAG spoken dialogue LGM
Novak91, DobesNovak91 Novak, Dobes 1991 1991 German, English LILOG large-scale NLP system with a generation component large-scale NLP system with a generation component 2000-06-03 LILOG
Copestake02 Copestake 1996 ongoing independent LKB General workbench for typed feature grammar formalisms, including a tactical generation component The LKB system is a grammar and lexicon development environment for use with constraint-based linguistic formalisms. The LKB software is distributed by the LinGO initiative, a loosely-organized consortium of research groups working on unification-based grammars and processing schemes. The download page for the LKB system and other available LinGO software and grammars is http://lingo.stanford.edu/ftp. <P>The LinGO English Resource Grammar (with associated lexicons and so on) may be downloaded from http://lingo.stanford.edu/ftp as erg.tgz. Note that it requires at least 250 MBytes of RAM to run with the LKB. No documentation for the ERG is available at this time (other than the source file comments). <P>The LKB system includes a tactical generation capability.<P>The LKB system is Open Source and there is no registration process. 2002-04-28 http://www-csli.stanford.edu/~aac/lkb.html LKB
Block91 Block, Gehrke, Hunze, Schachtl, Schmid, Zünkler 1990 1993 German LKP the Linguistic Kernel Processor (LKP) offers parsing and generating in TUG the Linguistic Kernel Processor (LKP) offers parsing and generating in TUG 2000-06-03 http://www.dfki.de/lt/registry/parsers/lkp.html independent LKP
Smith-etal94 Garigliano, Smith, Morgan 1994 1994 English Lolita general NLP including generation general NLP including generation 2000-06-03 independent Lolita
McKeown-etal97-magic McKeown, Feiner, Jordan, Dalal, Pan, Shaw, Zhou, Chen, Concepcion 1996 1997 English MAGIC Multimedia Abstract Generation for Intensive Care data MAGIC (Multimedia Abstract Generation for Intensive Care data) is a real-time multimedia presentation system in the medical domain which generates concise and fluent spoken reports by incorporating prosody modeling for speech and advance linguistic constructions such as right-node raising and ellipsis. 2000-06-03 FUF medical FUF MAGIC
Goldberg-etal88 Goldberg, Kittredge, Polguère 1988 1988 English MARWORDS Marine weather forecast report generator (another precursor to FoG). Marine weather forecast report generator (another precursor to FoG). 2000-06-20 weather MARWORDS
Buchanan-etal92 Buchanan, Moore, Foresythe, Banks, Ohlsson 1992 1992 English MIGRAINE health explanations health explanations 2000-06-20 medical MIGRAINE
Tanaka-Ishii-etal98 Tanaka-Ishii, Hasida, Noda 1998 1998 English, French, Japanese MIKE automatic soccer commentary system MIKE is an automatic commentary system that generates a commentary of a simulated soccer game in English, French and Japanese 2000-06-20 soccer MIKE
Frenger98 Murray 1998 1998 English Mind.Forth Verbal output of a general artificial intelligence. Verbal output of a general artificial intelligence. 2001-09-05 http://sourceforge.net/projects/mind/ Mind.Forth
Turner94 Turner 1984 1993 English MINSTREL A story generator A story generator 2001-12-05 story generation MINSTREL
Callaway-etal98, Towns-etal98, Daniel-etal99 Callaway, Daniel, Lester, Bares, Towns 1998 2000 English, Spanish ml-CINESPEAK employs both natural language and generated 3D-graphical visualisations and animations A multilingual generation component within a learning environment employing both natural language and generated 3D-graphical visualisations and animations 2001-06-01 FUF learning FUF ml-CINESPEAK
Yao-etal98 Yao, Zhang, Wang 1998 1998 Chinese MLWFA multilingual weather forecast generation multilingual weather forecast generation 2000-06-03 weather MLWFA
Lavoie-etal97, Lavoie-etal96 Lavoie, Rambow, Reiter 1997 1997 English ModelExplainer produces textual descriptions of object-oriented software models produces textual descriptions of object-oriented software models 2001-09-05 http://www.cogentex.com/research/modex/index.shtml software ModelExplainer
Stede99-book, Stede98-cl, Stede96-enlgw, Stede96-phd Stede 1996 1996 English, German MOOSE multilingual paraphrases from a KL-ONE-representation; domain: technical manuals, emphasis on lexicalization multilingual paraphrases from a KL-ONE-representation; domain: technical manuals, emphasis on lexicalization 2000-06-20 manuals Penman, KPML MOOSE
Androutsopoulos-etal01, Dimitromanolaki-etal01, Karamanis-etal-2004-inlg, Isard-etal2003 Androutsopoulos, Kokkinaki, Dimitromanolaki, Calder, Oberlander, Not , O'Donnell 2001 2003 English, Greek, Italian, Spanish M-PIRO Development of technology for allowing museums to generate automatically textual or spoken descriptions of exhibits for collections across the Web or in virtual reality environments. Development of technology for allowing museums to generate automatically textual or spoken descriptions of exhibits for collections across the Web or in virtual reality environments. M-PIRO stands for "multilingual personalised information objects". 2002-09-15 SFG http://www.ltg.ed.ac.uk/mpiro/ museums M-PIRO
Spyns-etal97 Spyns, Deprez, Van Tichelen, Van Coile 1995 1998 English MTS A message-to-speech system for traffic message control systems A message-to-speech system for traffic message control systems 2001-11-10 spoken dialogue MTS
Zeng95-pacling, Matthiessen-etal95, Zeng93, Zeng96-phd Zeng 1993 1995 English, Chinese MulTex multilingual generation of news reports multilingual generation of news reports 2000-06-03 SFG news MulTex
Coch98-multimeteo, Coch98-inlg, Coch-etal99, Coch99-gat Coch, Chevreau 1998 1999 English, French, Spanish, German, Dutch Multimeteo in-use (interactive and automatic) generation of weather forecasts in-use (interactive and automatic) generation of weather forecasts. URL of the operational site is <A href = http://www.inm.es/wwi/MultiMeteo/Multimeteo.html>http://www.inm.es/wwi/MultiMeteo/Multimeteo.html</A>. 2000-06-03 http://cordis.europa.eu/data/MSS_PROJ_FP4_FR/ACTIONeqDndSESSIONeq25780200595ndDOCeq6ndTBLeqEN_PROJ.htm weather Multimeteo
McDonald80, Meteer-etal87 McDonald, Meteer 1975 1990 English MUMBLE surface generator (MUMBLE-86 being portable) surface generator (MUMBLE-86 being portable) 2000-01-01 independent MUMBLE
Novak:1986, Novak:1988-book, Novak:1987, NeumannNovak86 Novak 1985 1987 German NAOS generation of coherent scene descriptions generation of coherent scene descriptions 2000-01-01 scene description NAOS
Becker-etal92 Strobel 1992 1992 German NAUDA sentence generation in the context of database interfaces sentence generation in the context of database interfaces 2000-06-03 http://www.dfki.de/lt/registry/generation/nauda.html NAUDA
KnightHatzivassiloglou95, LangkildeKnight98 Knight, Hatzivassiloglou, Langkilde 1995 1995 English Nitrogen Statistical generation from interlinguas using a lattice intermediate representation and bigram statistical model Statistical generation from interlinguas using a lattice intermediate representation and bigram statistical model 2000-06-03 http://www.isi.edu/natural-language/projects/nlg-publications.html Nitrogen
FroelichRiet97, Froelich98 Frölich 1995 1998 English NLG(db) Generation of NL descriptions from information maintained in a conceptual database modelling language Generation of NL descriptions from information maintained in a conceptual database modelling language 2000-06-20 SFG KPML NLG-DB
RubinoffFainLehman94 Rubinoff, Lehman 1994 1994 English NL-Soar generation within the general AI system SOAR generation within the general AI system SOAR 2000-06-03 NL-Soar
Jablonski88 Jablonski et al. 1988 1990 German NUGGET text generation from a linguistic basis text generation from a linguistic basis 2000-01-01 NUGGET
Aguado-etal98, GalindoAguado01 Aguado, Gómez-Pérez, Bateman 1997 1999 Spanish Ontogeneration generation of database and ontology definition descriptions generation of database and ontology definition descriptions 2000-06-03 SFG ontology KPML Ontogeneration
deRosis-etal99 de Rosis, Grasso, Berry 1991 1999 English OPADE a 'watch-dog' decision support system for medical care containing a generation component for instructional texts a 'watch-dog' decision support system for medical care containing a generation component for instructional texts 2000-06-03 medical OPADE
Cohen78, Cohen81 Cohen 1978 1981 English OSCAR speech act planning and planning of the message (no linguistic output) speech act planning and planning of the message (no linguistic output) 2000-01-01 OSCAR
vanZanten-etal98, Nederhof-etal97 van Noord, Veldhuijzen van Zanten, Sima'an, Bonnema 1996 2000 Dutch OVIS A dialogue system for providing information about public transportation (OVIS: Openbar Vervoer Informatie System) A dialogue system for providing information about public transportation (OVIS: Openbar Vervoer Informatie System) 2001-11-02 spoken dialogue OVIS
Swartout77 Swartout 1977 1977 English OWL Generates explanations of how program produced advice by translating code to English Generates explanations of how program produced advice by translating code to English 2000-06-03 online help OWL
Fiedler01-ijcai, Fiedler01-ijcar, Fiedler01-phd, Fiedler99 Fiedler 2000 ongoing English P.rex an interactive, user-adaptive proof explanation system P.rex is an interactive proof explanation system that adapts its explanations to the user and flexibly reacts to his questions or requests.<P><H2>Architecture</H2> <P><IMG SRC=http://www.ags.uni-sb.de/~prex/figures/architecture.jpg> <P>As a generic system, P.rex can be connected to different theorem provers, namely by means of the formal language TWEGA for specifying proofs and mathematical theories. Mathematical theories are organized in a hierarchical knowledge base. Each theory in it may contain, for example, axioms, definitions, and theorems along with proofs. A proof of a theorem can be represented hierarchically in TWEGA such that the various levels of abstraction are made explicit. <P>The central component of the system is the dialog planner. It is implemented in ACT-R, a goal-directed production system that aims to model human cognition. In ACT-R, declarative and procedural representations of knowledge are explicitly separated into the declarative memory and the procedural production rule base. The plan operators of the dialog planner are defined in terms of productions and the discourse plan is represented in the declarative memory. <P>The dialog plan is passed on through the sentence planner to the surface realizer. Currently, we use a derivate of PROVERB's sentence planner to plan the internal structure of the sentences, which are then realized by the surface realizer TAG-GEN. <P> The uttered sentences are finally displayed on the user interface. It also allows the user to enter remarks, requests and questions anytime. An analyzer receives the user's interactions and passes them on to the dialog planner. In the current stage, we use a simplistic analyzer that understands a small set of predefined interactions. 2002-04-28 http://www.ags.uni-sb.de/~prex proofs PROVERB, TAG-GEN P.rex
Finney-etal:1987 de Finney et al. 1987 1987 French PARDA generation of arguments System for the generation of arguments. 2000-01-01 arguments PARDA
Ebert-etal96, Porzel-etal2002 Ebert, Jansche, Klabunde, Porzel 1996 1996 German PAROLE generates addressee-oriented spatial descriptions (partner-oriented linguistic explicating): The system generates spatial descriptions. It is an example for cognitive modeling, because the choice which point of view the system should take to determine spatial directions is based on several empirical studies on addressee-orientation in speaking. Furthermore, the system generates certain hypotheses on viewpoint-taking if some features of the addressee (like his/her position within the spatial environment, his cognitive competence etc.) are known. 2000-06-20 spatial descriptions PAROLE
Colby75 Colby 1975 1975 English Parry simulation of paranoid processes simulation of paranoid processes 2000-06-20 psychiatry Parry
Sheremetyeva-etal96, SheremetyevaNirenburg96 Sheremetyeva, Nirenburg 1995 1998 English Patent Claim Expert generates patent claims generates patent claims concerning devices on the basis of knowledge elicitation where the user selects objects and relations from an ontology 2000-06-03 patents Patent Claim Expert
Granville83 Granville 1983 1983 English PAUL generation of coherent texts generation of coherent texts 2000-01-01 PAUL
Hovy88-book, Hovy87-JPrag Hovy 1986 1988 English PAULINE pragmatically motivated text planning pragmatically motivated text planning 2000-01-01 PAULINE
Moore89-thesis, MooreSwartout90-catalina Moore 1989 1994 English PEA generates a system's dialogue contributions teaching a student about a programming language generates a system's dialogue contributions teaching a student about a programming language 2000-06-03 SFG tutoring Penman PEA
Milosavljevic99-phd, MilosavljevicDale96, Dale-etal98, MilosavljevicOberlander98, Milosavljevic97, Milosavljevic-etal96 Milosavljevic, Dale 1996 1996 English Peba-II produces web pages that describe or compare animals, with embedded hypertext links Peba-II is an on-line animal encyclopedia which produces descriptions and comparisons of animals as world wide web pages. The Peba system contains information about hundreds of animals, but it has more information about Australian Monotremes (the Echidna and Platypus) and Marsupials (such as the Kangaroo, Koala and Wombat). There is also a lot of information on the cat family (Lion, Cheetah, Tiger, Leopard, Snow Leopard etc), and lots of other mammals like the Alligator, Porcupines (American and African), Elephants (African and Indian), Rabbit, Baboon, Gorilla, Plains Zebra, Goat, Sheep, and so on.<P>The system produces descriptions and comparisons of animals by using underlying representation of information, similar to a database. The web pages you are given do not exist on the server, but are dynamically created as and when you ask for information. Because the descriptions are created by a system, the system can adapt the descriptions to your particular knowledge or experience. The system thus produces web pages that describe or compare animals, with embedded hypertext links. 2001-08-04 http://www.dynamicmultimedia.com.au/peba/ animal comparisons Peba-II
Bullock94-msc Bullock 1994 1994 English Pen&Pad Reporter Describes a NLG scheme for use with the Pen&Pad patient care workstation system Describes a NLG scheme for use with the Pen&Pad patient care workstation system, devised by the Medical Informatics Group at the University of Manchester. The system produces texts describing the contents of the medical records. 2000-06-20 medical Pen&Pad Reporter
MatthiessenBateman91, MannMatthiessen85-benson, Hovy-etal92-nlgw, Hovy93-AI, Matthiessen90-coling, MannMatthiessen83, Kasper89-spl, Matthiessen88-inquiries Mann, Matthiessen, Hovy, Kasper, Bateman, Whitney, Albano, Poulton, O'Donnell 1981 1992 English PENMAN Portable systemic generator for domain-independent large-scale text generation Penman was the major text generation project established by William C. Mann in the early 1980s. Within this project were developed resources such as the very large Nigel grammar of English and several approaches to text planning that came to define some of the standard approaches in the field. 2000-01-01 SFG PENMAN
Finkler97-phdbook Finkler 1997 1997 German PERFECTION Performing Self-Corrections During Incremental Generation of Natural Language Performing Self-Corrections During Incremental Generation of Natural Language 2000-06-20 PERFECTION
Jacobs85 Jacobs 1985 1985 English PHRED multilingual generator multilingual generator 2000-01-01 PHRED
Beale96 Beale 1996 1996 English Picard sentence generation sentence generation 2000-06-03 Picard
Binsted-etal95, Cawsey-etal96-ecai, Cawsey-etal95-dagstuhl Cawsey, Binsted, Bental, Jones, Carter, Grasso 1993 1998 English PIGLIT personalized hypertext explanations of on-line patient medical records PIGLIT (personalized patient education): a system for generating personalized hypertext explanations of on-line patient medical records so as to help patients understand them 2000-06-03 http://www.cee.hw.ac.uk/~diana/piglit/pigtext.html medical PIGLIT
KreyssNovak90 Kreyss, Novak 1990 1990 German PIT text planning text planning 2000-01-01 PIT
McKeown-etal94-plandoc McKeown, Kukich, Shaw 1994 1994 English PLANDOC summary generation reporting changes in a telephone network summary generation reporting changes in a telephone network 2000-01-01 FUF telecom FUF PLANDOC
Waltz, Waltz76-PLANES Waltz, Finin, Green, Conrad, Goodman, Hadden 1975 1978 English PLANES database access and retrieval database access and retrieval 2000-06-20 PLANES
Neumann, van Noord 1994 1994 German, Dutch PlayMoBild bidirectional linguistic deduction, parsing and generation of single sentences bidirectional linguistic deduction, parsing and generation of single sentences 2000-06-03 http://www.dfki.de/lt/registry/parsers/playmobild.html PlayMoBild
Beurer91, Zadrozny93 Beurer 1990 1993 English PLNLP general NLP including generation general NLP including generation 2000-06-03 independent PLNLP
Emele-etal90-coling Momma, Zajac, Emele 1990 1992 English, German Polygloss typed-unification-based generation Typed-unification-based generation system 2000-06-03 TFS independent Polygloss
FinklerNeumann89, Finkler89 Finkler, Neumann 1989 1989 German POPEL-HOW incremental sentence generation incremental sentence generation 2000-01-01 POPEL-HOW
Reithinger91, Reithinger87-gwai Reithinger 1987 1991 German POPEL-WHAT incremental sentence generation incremental sentence generation 2000-01-01 POPEL-WHAT
FascianoLapalme2000, FascianoLapalme96 Fasciano, Lapalme 1996 1996 French PostGraphe combined text and graphics generation combined text and graphics generation 2000-06-03 PostGraphe
Dale-etal98-realities, Milosavljevic-etal98-highway Green, Dale, Williams 1998 1998 English Power another virtual museum system dynamically generating descriptions of museum objects Another virtual museum system dynamically generating descriptions of museum objects. Original prototype based on Peba-II. Final system closely resembling ILEX in output although more like Peba in architecture. 2000-06-03 template http://www.mri.mq.edu.au/~peba/Power/system.html museums Peba-II Power
Andre-etal98-iui, Andre-etal93-ppp André, Müller, Rist 1993 1996 English, German PPP creates an animated character which explains to a user how to perform various tasks, using a mixture of text, speech and graphics creates an animated character which explains to a user how to perform various tasks, using a mixture of text, speech and graphics 2000-06-03 TAG instructions PPP
Gulla96, Gulla93, GullaWillumsen93, Gulla-etal91 Gulla 1991 1994 English PPP An explanation generation facility for knowledge validation in computer-aided software engineering A system that uses rhetorical strategies for the generation of explanation of dynamic and other information within a CASE (computer-assisted software engineering) system, all collectively called PPP. 2001-09-18 FG CASE models PPP-case
Jameson-etal94 Jameson 1994 1994 German PRACMA dialogue system with user modelling dialogue system with user modelling 2000-06-20 PRACMA
Andre95-book André 1995 1995 German PREPLAN plan-based multimedia generation plan-based multimedia generation 2000-01-01 TAG PREPLAN
Lavoie98 Lavoie, Rambow 1998 1998 English Presentor A Framework for Customizable Generation of Hypertext Presentations A Framework for Customizable Generation of Hypertext Presentations 2000-06-03 Presentor
GagnonLapalme96 Gagnon, Lapalme 1995 1995 French PRETEXTE generation of temporal expressions generation of temporal expressions 2000-01-01 SFG PRETEXTE
Dik96, Dik92 Dik 1990 1996 English, French, Dutch ProfGlot sentence generation with Functional Grammar sentence generation with Functional Grammar 2000-06-03 FG ProfGlot
WhiteCaldwell98 Caldwell 1997 1999 English Project Reporter generates dynamic web-based project status reports generates dynamic web-based project status reports from files created with Microsoft Project or other compatible project management software. Reports feature hyperlinked textual descriptions of project elements, as well as coordinated multimodal display with an interactive GANNT chart applet. Commercially available over the web since 3/99. 2000-06-03 http://www.cogentex.com/products/reporter/ Commercial Project Reporter
Adorni87 Adorni 1987 1987 English Promethius sentence generation sentence generation 2000-06-03 Promethius
MillerRennels88 Miller, Rennels 1988 1988 English PROSENET 2000-01-01 PROSENET
Nicolov-etal96, NicolovMellish99-englw, Nicolov-etal97-ranlp Nicolov et al. 1995 1999 English PROTECTOR flexible generation of sentences from conceptual graphs flexible generation of sentences from conceptual graphs 2000-01-01 DTG PROTECTOR
Davey78 Davey 1978 1978 English PROTEUS generation of comments when playing TIC-TAC-TOE generation of comments when playing TIC-TAC-TOE 2000-01-01 SFG game commentary PROTEUS
HuangFiedler97, HuangFiedler96 Fiedler, Huang 1996 2000 English PROVERB natural language presention of machine-found proofs PROVERB is a system presenting machine-found proofs in natural language. It is embedded in the interactive proof development environment OMEGA. Settled between automated theorem proving (ATP) and natural language generation (NLG), PROVERB uses techniques of both these fields. In order to present proofs in a way as they can be found in mathematical textbooks, PROVERB first transforms and abstracts resolution proofs or natural deduction proofs to so-called assertion level proofs, which abstract from the basic calculus inference rules. Here, most justifications are in terms of the application of a definition, a lemma, or a theorem, which are collectively called assertions. Assertion level proofs are then verbalized in English using NLG techniques. First, a macroplanner determines the order and content of the information to be conveyed. Then, a microplanner plans the internal structure of the paragraphs and sentences. Finally, the surface sentences are produced by the system TAG-GEN. Furthermore, PROVERB allows for the automatic creation of a LaTeX document from the English proof. 2002-04-28 http://www.ags.uni-sb.de/~afiedler/ formal proofs TAG-GEN PROVERB
Lehnert78-book Lehnert 1977 1978 English QUALM goal sensitive elaboration of a response (the depth depending on the nature of the question) goal sensitive elaboration of a response (the depth depending on the nature of the question) 2000-01-01 QUALM
Cahill-etal00-rags, Cahill-etal01, RAGS-intro-01, Cahill-etal99, rags99 Mellish, Scott, Evans, Doran, Tipper, Cahill, Paiva, Reape 1997 1999 independent RAGS aims to develop a generic architecture for applied NLG systems aims to develop a generic architecture for applied NLG systems 2001-11-03 http://www.itri.brighton.ac.uk/projects/rags/ RAGS
Kittredge-etal86-coling Kittredge et al. 1986 1986 French, English RAREAS generation of weather forecast reports (a precursor to FoG) generation of weather forecast reports (a precursor to FoG) 2000-01-01 MTM weather RAREAS
LavoieRambow97-realpro Lavoie, Rambow 1997 1997 English RealPro a syntactic realisation system based on meaning-text theory a syntactic realisation system based on meaning-text theory 2000-06-03 MTM independent RealPro
Codd, Codd77 Codd 1976 1977 English RENDEZVOUS database query formulation system database query formulation system 2000-06-20 DB RENDEZVOUS
Cahill-etal01 Cahill, Evans, Paiva, Scott, Power, Carroll, van Deemter 1999 2000 English RICHES Instructional texts for patients Generates instructional texts for using different medicine delivery methods from rhetorical plans 2001-09-18 instructions RICHES
Claassen97 Claassen 1996 1997 Dutch ROIS Medical domain generation medical domain generation 2001-09-18 segment grammar medical ROIS
McCoy86 McCoy 1985 1986 English ROMPER correction and anticipation of wrong inferences correction and anticipation of wrong inferences 2000-01-01 ROMPER
Harvey98 Harvey, Carberry 1998 1998 English RTPI integration and aggregation of RST-style trees for generating medical critiques RTPI (Rule-based Text Plan Integrator): integration, aggregation of RST-style trees for generating medical critiques 2000-06-03 http://xxx.lanl.gov/abs/cmp-lg/9805010 medical RTPI
RothHefley93 Roth, Hefley 1993 1995 English SAGE multimodal explanations multimodal explanations 2000-06-03 explanation FUF SAGE
Sibun91, Sibun91-phd Sibun 1991 1991 English Salix generates extended texts in domains including house layouts and family relations. Salix incrementally generates extended texts whose structure is derived from the domain structure of the subject matter, in several domains including house layouts and family relations. The code still works. 2000-06-03 house layouts Salix
SCHOLAR Carbonell et al. 1973 1973 English SCHOLAR natural language interface for teaching geography natural language interface for teaching geography 2000-01-01 teaching geography SCHOLAR
Verret-etal99 Vigneux, Marcoux, Verret 1991 2000 French, English Scribe Weather forecast report generator (template-based). Weather forecast report generator (template-based). The system is deployed across Canada, but its acceptance and use is regionally dependent (same could be said for FoG). 2000-06-20 http://www.cmc.ec.gc.ca/cmc/bibliotheque/PREVISIONS/f_9_2.htm weather Scribe
AlexanderssonPoller98-inlg, AlexanderssonPoller2000-inlg, AlexanderssonPoller2000, Reithinger-etal2000 Alexandersson, Engel, Kipp, Poller, Reithinger 1997 2000 English, German Script generator of Verbmobil Generation of summaries and minutes for spoken multilingual negotiation dialogues Generation of summaries and minutes for spoken multilingual negotiation dialogues 2000-09-18 http://www.dfki.uni-sb.de/~janal summary Script generator of Verbmobil
Roesner88 Rösner 1986 1987 German SEMSYN summaries about the labor market summaries about the labor market 2000-01-01 labor market SEMSYN
Roesner87 Rösner 1987 1987 German SemTex Automated newsagency Automated newsagency 2000-06-03 news SEMSYN SemTex
Busemann95-sereal Busemann 1995 1995 German SeReal 2000-06-20 SeReal
Winograd72 Winograd 1972 1972 English SHRDLU description of a microworld (cubes, blocks) The pioneer system capable of answering questions and moving objects within a microworld consisting of variously coloured cubes, blocks, pyramids, and so on. The system answers questions, executes commands, and accepts information in an interactive English dialog. The system contains a parser, a recognition grammar of English, programs for semantic analysis, and a general problem solving system. It can remember and discuss its plans and actions as well as carrying them out. From the generation perspective the interest of the system lies in its dialog capabilities, particularly, for example, the ability to initiate clarification dialogs when references were not clear. Knowledge in the system is represented in the form of procedures, rather than tables of rules or lists of patterns. The system showed natural language, planning and actions working together for the first time in a complete system and would still give some current systems a run for their money! <P>A 'recreation' of SHRDLU has now been prepared by the <A HREF=http://www.semaphorecorp.com/misc/shrdlu.html>Semaphore corporation</A>. 2002-11-23 http://hci.stanford.edu/~winograd/shrdlu/ spatial relationships SHRDLU
Dilley-etal92 Dilley, Thiel, Tissen, Bateman 1992 1992 English SIC! integrated multimodal interaction to information systems integrated multimodal interaction to information systems 2000-06-03 SFG Penman SIC!
Driedger-etal2000 Driedger 1999 1999 English, French Siren Marine weather forecast report generator Marine weather forecast report generator 2000-06-20 weather Siren
Patten88-book, Patten86 Patten 1984 1986 English SLANG use of systemic grammars for NLG use of systemic grammars for NLG 2000-01-01 SFG SLANG
Wahlster2006, Wahlster01, Wahlster2003-fusion Wahlster et al. 2000 2006 German SmartKom A project to develop a multimodal, multimedia, self-explanatory, user-adaptive interface for the dialogic interaction of humans and technology. A project to develop a multimodal, multimedia, self-explanatory, user-adaptive interface for the dialogic interaction of humans and technology. This should include substantial presentation design components, involving natural language as one of several output modalities. <H2>Architecture</H2> <P><IMG SRC=http://smartkom.dfki.de/image/archi_eng.gif> 2008-01-10 http://smartkom.dfki.de/eng/start_en.html SmartKom
Herzog:1989, Andre-etal88-soccer Herzog, André, Rist, Bosch 1987 1989 German SOCCER generation of soccer reports generation of soccer reports 2000-01-01 soccer SOCCER
Hitzeman-etal99 Hitzeman, Oberlander, Taylor 1997 1999 English SOLE the project aims to develop and evaluate a rhetorically-oriented interface between a natural language generation system and a speech synthesis system. the project aims to develop and evaluate a rhetorically-oriented interface between a natural language generation system and a speech synthesis system. 2000-06-03 http://www.cstr.ed.ac.uk/projects/sole.html SOLE
Teich-etal97-specom, Grote-etal96, Grote-etal97-speak, HagenStein96, Speak-description98, Stein98-speakeval Bateman, Stein, Hagen, Teich, Grote, Thiel 1995 1996 German SPEAK! spoken language generation for supporting database access spoken language generation for supporting database access 2000-06-03 SFG metadialog KPML SPEAK!
Kosseim-etal96, KosseimLapalme2000 Kosseim, Lapalme 1995 1995 French SPIN planning of instructional texts planning of instructional texts 2000-01-01 instructions SPIN
Pollack:1986 Pollack 1986 1986 English SPIRIT determination of the adequacy of information determination of the adequacy of information 2000-01-01 SPIRIT
Meteer92-anlp, Meteer90-catalina Meteer 1990 1992 English SPOKESMAN producing texts by revising producing texts by revising 2000-01-01 TAG MUMBLE SPOKESMAN
StoneDoran97, StoneDoran96, Stone-etal2000 Stone, Doran 1997 ongoing English SPUD sentence planning using descriptions sentence planning using descriptions 2000-06-03 TAG http://www.cs.rutgers.edu/~mdstone/nlg.html SPUD
ReiterDale2000 Dale et al. 1999 1999 English StockReporter produces descriptions of stock market data produces descriptions of stock market data: StockReporter takes numeric data that describes the performance of a particular stock produces from this data, given an user-specified Reporting Period, a textual summary that describes how the stock performed over that reporting period. 2000-06-03 http://www.mri.mq.edu.au/stockreporter/ stock market StockReporter
Reiter-etal03-ai, Lennox:bmj01, Reiter-etal97-content, Reiter-etal99, Reiter:cl00, Reiter:nlg00, Reiter:acl01, Reiter-etal97-ewnlg, ReiterOsman97 Reiter, Robertson, Osman, Lennox 1997 2001 English STOP Generates personalized smoking-cessation letters The STOP (Smoking Termination with cOmputerised Personalisation) system generates personalized smoking-cessation letters. The aims of the project are: <UL><LI>To develop a computer system for generating tailored letters to help people stop smoking<LI> To research knowledge acquisition (KA) techniques to acquire text-planning and sentence-planning rules from domain experts <LI>To evaluate the clinical effectiveness of the computer generated letters in a general practice setting<LI>To evaluate the cost effectiveness of this brief smoking cessation intervention</UL>The system evaluation has now been completed.<P>An online demo of the system can be reached <A href=http://www.csd.abdn.ac.uk/research/stop/onlineQ.htm>here.</A> 2001-09-05 http://www.csd.abdn.ac.uk/research/stop medical, smoking cessation STOP
CallawayLester95, CallawayLester97, Callaway01, Callaway00-phd, CallawayLester01-ijcai, CallawayLester01-cogsci Callaway, Lester 1994 2001 English, Spanish StoryBook Multi-page narratives in the Little Red Riding Hood domain StoryBook produces multi-page narratives in the Little Red Riding Hood domain at the deep structure level using the FUF/SURGE surface realizer. It follows a standard pipelined architecture for generating English and (to a lesser extent) Spanish. The English component includes revision, a discourse history, and limited lexical choice. StoryBook underwent a comprehensive evaluation in March 2000. 2001-09-26 FUF http://ecate.itc.it:1024/callaway/ little red-riding hood FUF, SURGE StoryBook
Robin96, Robin94-phd, Robin93 Robin, McKeown 1993 1996 English STREAK revision-based generation of summaries of baseball games revision-based generation of summaries of baseball games 2000-01-01 FUF baseball FUF STREAK
SaggionLapalme2000 Lapalme, Saggion 1999 2000 English Summum Generates indicative and informative abstracts in English Generates indicative and informative abstracts in English 2000-06-20 abstracting Summum
Sripada-etal01-ka, Sripada-etal01-weather, Yu-etal2002-gas, Yu-etal01-sumtime, ReiterSripada2002-inlg, Sripada-etal2002-corpus, ReiterSripada-fc-cl, Sripada-etal2002-weather, Sripada:enlg01, ReiterSripada-2004-inlg, Reiter2007-data-to-text Reiter, Sripada, Yu, Hunter 2001 ongoing English SumTime English summary descriptions of time-series data sets Our goal is to develop technology for producing English summary descriptions of a time-series data set. Currently there are many visualisation tools for time-series data, but techniques for producing textual descriptions of time-series data are much less developed. ... <P>From a natural-language generation perspective, we expect to focus on lexicalisation and user-modelling. For example, we need to write rules which determine when a particular data set can be described as "rising"; prototype or exemplar models of word meaning may be useful here. We also need to define user models which allow us to decide what information should be communicated in a space-limited text, given the huge number of things that could be said about a particular data set. This should be a very exciting project for anyone interested in the challenge of generating good English texts from non-linguistic input data.<P>To take a concrete example, consider the following extract from a (real-life) weather report: <blockquote>Saturday will be yet another generally dull day with early morning mist or fog and mainly cloudy skies being prevalent. There will be the odd bright spell here and there, but it will feel rather damp with patches of mainly light rain to be found across many parts, especially the west and south.</blockquote> This was produced by a human forecaster from an underlying data set (generated by numerical modelling and prediction techniques) which gave predictions for meteorological parameters such as precipitation, temperature, wind speed, and cloud cover at various altitudes, at regular intervals for various points in the area of interest. The underlined phrases are some of the places where the human forecaster has used short textual summaries to describe this complex data. Our goal is to be able to perform similar types of summarisation in a computer system. That is, for example, to take raw data on cloud cover and solar flux over course of a day, and from this produce a phrase such as odd bright spell here and there. This involves research in both time-series analysis (to detect patterns and phenomena in the data) and natural-language generation (to decide how to best describe these patterns and phenomena in language). 2002-10-12 http://www.csd.abdn.ac.uk/research/sumtime weather, gas turbines SumTime
YoudMcGlashan92 McGlashan 1990 1992 English SUNDIAL spoken language dialogue system spoken language dialogue system 2000-06-20 spoken dialogue SUNDIAL
Kraus02, Hueske-Kraus2003 Kraus 2002 ongoing German SUREGEN-2 Medical document generation A system for the generation of medical documents such as findings, surgical reports or referral letters. SUREGEN-2 uses a hybrid approach, combining canned text phrases (with variables) and bottom up generation. It includes<UL><LI>a (simple, though sufficient) surface generator<LI> a predefined, rudimentary but open ontology for medicine<LI> mechanisms for conceptual and sentence aggregation<LI> a mechanism for referring expressions<LI> a feature for the quite frequent (in medicine) enumerative phrases<LI> a module for generation based on semantic functions (recurring semantic constellations)<LI>a simple mechanism for dealing with synonymity, antonymity and cohyponyms, mainly used for aggregation</UL><P>A standalone demo version of the system is available for download <A HREF=http://www.suregen.de/ftp/standalone1.zip>here</A>. The demo requires some facility with Common Lisp. The system will be used within a larger application context and will be in 'part of a larger enterprise which will be in routine use in our hospital by DEC 2002' (Dirk Kraus, email, 4.Nov.02). 2008-04-18 http://www.suregen.de medical SUREGEN-2
ElhadadRobin96 Elhadad, Robin 1994 1998 English SURGE the English grammar used with FUF the English grammar used with FUF 2000-06-03 FUF http://www.cs.bgu.ac.il/research/projects/surge/index.htm independent FUF SURGE
Luckhardt83 Luckhardt 1983 1983 English, French, Dutch Susy MT system with some attention to generation MT system with some attention to generation 2000-06-03 Susy
Busemann88, Busemann:1987 Busemann 1986 1988 German SUTRA surface generator of HAM-ANS dialog system surface generator of HAM-ANS dialog system 2000-01-01 GPSG reservations SUTRA
Zock89 Zock 1989 1991 French SWIM CALL system for learning to speak French CALL system for learning to speak French 2000-01-01 teaching French SWIM
Nogier91 Nogier 1991 1991 French SYLAE sentence generation from conceptual graphs sentence generation from conceptual graphs 2000-01-01 SYLAE
Abb-etal93, Abb-etal96-synphonics, Guenter-etal95 Abb, Günther, Herweg, Lebeth, Maienborn, Schopp 1993 1996 German Synphonics Spoken dialogue contribution generation Spoken dialogue contribution generation 2000-06-03 spoken dialogue Synphonics
Harbusch-etal91 Finkler, Harbusch, Kilger 1990 1993 German, English TAG-GEN syntactic generator exploiting an incremental and parallel processing scheme syntactic generator exploiting an incremental and parallel processing scheme; used in WIP and EFFENDI 2000-06-03 TAG http://www.dfki.de/lt/registry/generation/tag-gen.html TAG-GEN
Paris93, Cecile-user-CL88, CECILE-STRATEGIES Paris 1985 1988 English TAILOR tailoring the text to the user's knowledge tailoring the text to the user's knowledge 2000-01-01 FUG user modelling TAILOR
talespin, Meehan80 Meehan 1977 1977 English TALE-SPIN planning and generation of fables planning and generation of fables 2000-01-01 stories TALE-SPIN
RoesnerStede94-coling, RoesnerStede92-techdoc-pub Rösner, Stede, Grote 1992 1992 German TECHDOC multilingual instructions for automobile maintenance multilingual instructions for automobile maintenance 2000-01-01 SFG instructions Penman, SEMSYN TECHDOC
Appelt82-acl Appelt 1980 1982 English TELEGRAM planning for generation planning for generation 2000-06-20 TELEGRAM
Klenner91 Klenner 1991 1991 German TELOS temporal connective generation temporal connective generation 2001-09-20 temporal connectives TELOS
Schilder93 Schilder 1993 1993 German TEMPLOG temporal connective generation temporal connective generation 2000-06-03 temporal connectives TEMPLOG
Bus-Hor:97 Busemann, Horacek 1997 1998 German, French, English, Portuguese TEMSIS provides environmental information provides environmental information 2000-06-03 Template http://www.dfki.de/service/nlg-demo environment TG/2 TEMSIS
Bunt87 Bunt 1987 1987 English TENDUM generating with pragmatic information generating with pragmatic information 2000-06-20 TENDUM
MillerRennels88 Miller, Rennels 1988 1988 English TEXNET generation of clinical summaries generation of clinical summaries 2001-09-17 medical TEXNET
 ?? Maybury  ??  ?? English TEXPLAN  ??  ?? 2000-06-20 TEXPLAN
McKeown82-ACL McKeown 1980 1985 English TEXT use of schemata for determining content & text structure use of schemata for determining content and text structure 2000-01-01 TEXT
Buchin99 Buchin, Schmerl 1998 1999 German TextPro 2001-01-01 Template TextPro
Busemann96, Busemann98-bremen, Busemann99, Wein96 Busemann 1998 ongoing independent TG/2 shallow generation component using generalized templates TG/2 stands for a new generation of template-based generators. The system is organized as a classical production system, separating the generation rules from their interpreter. Generation rules are defined as condition-action pairs. They can represent not only templates but also context-free grammar rules, using category symbols. Integrating these rule types into a single formalism allows for a shallow modelling of language where this proves sufficient, or for more fine-grained models whenever necessary. The interpreter consists of the standard three-step processing cycle: determine the set of applicable rules, select one from this set, and execute it. The TG/2 system is used in the generators TEMSIS and COSMA. <P><H3>Architecture</H3><IMG SRC=../images/tg2-e.gif><P> 2001-09-06 Template http://www.dfki.de/~busemann/more-tg2.html independent TG/2
EdgarPelletier93 Edgar, Pelletier 1993 1993 English THINKER explanation of natural deduction proofs explanation of natural deduction proofs 2000-06-20 formal proofs THINKER
Mueller98 Mueller 1994 2000 English, French ThoughtTreasure generator Generates any ThoughtTreasure assertion for which appropriate lexical entries are defined Generates any ThoughtTreasure assertion for which appropriate lexical entries are defined. Under development since 1994, ThoughtTreasure is a comprehensive platform for natural language processing (English and French) and commonsense reasoning. ThoughtTreasure contains a database of 25, 000 concepts organized into a hierarchy. For example, <tt>Evian</tt> is a type of <tt>flat-water</tt>, which is a type of <tt>drinking-water</tt>, which is a type of <tt>beverage</tt>, which is a type of <tt>food</tt>, and so on. <p> Each concept has one or more English and French approximate synonyms, for a total of 55, 000 words and phrases. For example, associated with the <tt>food</tt> concept are the English words <em>food</em> and <em>foodstuffs</em> and the French words <em>aliment</em> and <em>nourriture</em> (and others). <p> ThoughtTreasure contains 50, 000 assertions about concepts, such as: a <tt>green-pea</tt> is a <tt>seed-vegetable</tt>, a <tt>green-pea</tt> is <tt>green</tt>, a <tt>green-pea</tt> is part of a <tt>pod-of-peas</tt>, and a <tt>pod-of-peas</tt> is found in a typical grocery store. <p> ThoughtTreasure contains about 100 scripts, or computer-understandable descriptions of typical activities such as going to a restaurant or birthday party. <p> ThoughtTreasure contains 70, 000 lines of C code implementing: <ul> <li> text agents for recognizing words, phrases, names, and phone numbers, <li> mechanisms for learning new words, <li> a syntactic parser, <li> a natural language generator, <li> a semantic parser for producing a surface-level understanding of a sentence, <li> an anaphoric parser for resolving pronouns, <li> planning agents for achieving goals on behalf of simulated actors, and <li> understanding agents for producing a more detailed understanding of a discourse. </ul> <P>The ThoughtTreasure system was developed by Erik Mueller under the guide of the company Signiform. Signiform was founded in 1997; it shut down business in 2000. 2001-12-05 http://www.signiform.com/tt/htm/tt.htm ThoughtTreasure
Paris-etal01-um, Lu-etal2000-tiddler 1999 ongoing English Tiddler Personalized travel guides Tiddler is a prototype system for delivering tailored information to users. Tiddler combines structured information from databases with text from existing web pages to deliver information tailored both to the user and the delivery medium. <P> The information is tailored for delivery to a full-featured web browser or a hand held device like a personal organiser. As screen size decreases, there is a need to reformat information while preserving meaning and clarity. Tiddler is designed to automate this process. Tiddler technology is applicable in any situation where tailoring of information from diverse sources is required to deliver value-added services to customers 2001-08-31 Template http://www.ted.cmis.csiro.au/proj/Tiddler/ tourism Tiddler
Cawsey-etal2000 Cawsey, Bental, Jones 1999 1999 English TIGGER Time-based Explanation Generation includes an explicit psychological model of patients' information needs. TIGGER (Time-based Explanation Generation Gives Explanatory Relevance) is an extension over PIGLIT that includes an explicit psychological model of patients' information needs. 2000-06-20 http://www.cee.hw.ac.uk/~diana/piglit/nlgtalk.html medical PIGLIT TIGGER
2001 2002 English TOPCAT Produces descriptions of malt whiskies Produces descriptions of malt whiskies using the same technology as the ILEX museum explorer 2002-10-13 http://www.hcrc.ed.ac.uk/ilex/demos/whisky/Web/ whisky ILEX TOPCAT
Webber-etal97 Webber, Carberry, Clarke, Gertner, Harvey, Rymon, Washington 1996 1999 English TraumGen Communication of information in clinical environments and for medical support systems Communication of information in clinical environments and for medical support systems 2001-11-02 medical TraumGen
Somers-etal97-tree Somers 1997 1997 English, Greek TREE multilingual generation of job advertisements multilingual generation of job advertisements 2000-06-03 job adverts TREE
DorrOlsen96, DorrGaasterland94 Dorr, Olsen, Gaasterland 1992 1996 English, Spanish Unitran MT system with some attention to generation MT system with some attention to generation 2000-06-03 LCS Unitran
 ??? Schütz, Becker 1996 ongoing? EU UNL An international cooperation to provide information on the internet in the languages of the members of the United Nations The UNL project is a large scale international cooperation with the goal to provide information on the internet in all national languages of the members of the United Nations. The United Nations University (UNU) in Tokyo which leads the project has developed a single representation language ``Universal Networking Language (UNL). Source language specific analysis components will automatically transfer texts into a UNL representation from which target language specific generation components can generate text in all languages. These generation components will be available as plug-ins for WWW-browsers. This approach reduces the enormous translation task to a one-time analysis into a UNL document and the subsequent utilization of a generator for the selected target language. 2000-06-03 http://www.dfki.de/services/NLG/IUI-generation/unl-e.html UNL
MaierReithinger95, Reithinger-etal95, Maier-etal97-clarif, Alexandersson-etal97-vm197, Alexandersson-etal98-vm226 Maier, Reithinger 1995 1998 German VERBMOBIL:dialog dialogue management within the Verbmobil translation system dialogue management within the Verbmobil translation system 2000-06-20 Verbmobil-dialog
Buchberger-etal91-dialog Buchberger 1991 1991 German VIE-DU dialogue generation dialogue generation 2000-06-20 VIE-DU
BuchbergerHoracek88 Buchberger, Horacek 1984 1988 German VIE-GEN sentence generation from a semantic network sentence generation from a semantic network 2000-01-01 VIE-GEN
Dalianis95-anldb Dalianis 1995 1995 English VINST database retrieval result aggregation database retrieval result aggregation 2000-06-03 VINST
HerzogWazinski94, Herzog:1989, Herzog-etal96, Herzog-etal98, Wahlster89-words, Herzog-etal93-vitra-conf, HerzogRist88 Herzog, Blocher, Gapp, Stopp, Wahlster, André, Rist, Nagel, Enkelmann, Zimmermann, Wazinski 1989 1999 German VITRA Generation of dynamically changing scene descriptions Generation of dynamically changing scene descriptions. VITRA: Visual Translator 2002-01-04 TAG VITRA
 ?? Finkler 1993 1996 English VM-GECO VerbMobil Generation Components (I) The generation task is solved by two modules. The microplanner decides about syntactic structures and word choice using constraint-satisfaction techniques. The syntactic generator uses a declarative grammar to produce a syntactically correct utterance. The generator's output is an (annotated) English utterance which is handed over to the speech synthesis component for articulation. The input to VM-GECO is a sequence of semantic representations for utterances in the target language. They have been produced by the transfer component of the VERBMOBIL system by mapping semantic representations for German or Japanese input utterances onto semantic structures for English. 2001-09-13 HPSG, TAG http://www.dfki.de/pas/f2w.cgi?iuic/geco-sol-e VM-GECO
 ?? Finkler 1997 2000 German, English, Japanese VM-GECO-II VerbMobil Generation Components (II) The generation component VM-GECO-II is based on declarative knowledge sources, such that the processing mechanisms can be used for several target languages. Just as in VM-GECO-I, the principle of off-line preprocessing is used to guarantee the consistency of knowledge sources of several generation components as well as to increase efficiency during on-line processing. For each target language an HPSG is used as basis for syntactic generation and is preprocessed off-line into a TAG. 2001-09-13 HPSG, TAG http://www.dfki.de/pas/f2w.cgi?iuic/vmII-e appointments VM-GECO-II
FinklerKilger95, Becker-etal98 Becker, Finkler, Karbusch, Kilger 1995 1995 German VM-GEM German surface generator for the Verbmobil project; incremental German surface generator for the Verbmobil project; incremental 2000-06-03 TAG VM-GEM
ODonnell95-wag, ODonnell94 O'Donnell 1990 1994 English WAG a system for developing and using systemic grammars for generation and analysis A system for developing and using systemic grammars for generation and analysis. The Sentence Generation component of this system generates single sentences from a semantic input. This semantic input could be supplied by a human user. Alternatively, the semantic input can be generated as the output of a multi-sentential text generation system, allowing such a system to use the WAG system as its realisation component. The sentence generator can thus be treated a black-box unit. Taking this approach, the designer of the multi-sentential generation system can focus on multi-sentential concerns without worrying about sentential issues. 2000-06-03 SFG http://www.wagsoft.com/wag.html independent WAG
Sigurd-etal92-weathra Sigurd, Willners, Eeg-Olofsson, Johansson 1992 1992 Swedish, English Weathra Weather forecast report generator Weather forecast report generator 2000-06-20 weather Weathra
Inui-etal92 Inui et al. 1992 1992 Japanese WEIVER revision-based architecture revision-based architecture 2000-01-01 WEIVER
Andre-etal93 Wahlster et al. 1993 1993 German WIP multimodal generation, instructions for technical products multimodal generation, instructions for technical products 2000-01-01 TAG instructions WIP
Horacek93, Horacek-etal88-wisber Horacek, Sprenger 1988 1993 German WISBER 2000-06-03 WISBER
Zuckerman90 Zukerman 1991 1991 English WISHFUL content planning for teaching algebra content planning for teaching algebra 2000-01-01 teaching algebra WISHFUL
ZuckermanMcConachy95 Zuckerman, McConachy 1995 1995 English WISHFULL-II explanation generation explanation generation 2000-06-20 explanation WISHFULL-II
PowerScott98, Power99-enlgw, Power99-eacl, Scott99-ijcai, Piwek-etal99, Kibble-etal99, Power-etal98, Scott-etal98, vanDeemterPower98 Power, Evans 1998 ongoing English WYSIWYM What you see is what you meant': knowledge-editing with natural language generation WYSIWYM aims to allow domain experts to encode their knowledge directly, by interacting with a feedback text, generated by the system, which presents the knowledge defined so far and the options for extending or revising it. Previous knowledge editors have provided a graphical interface so that users can interact with diagrams rather than writing code; WYSIWYM takes a step further, exploiting automatic text generation so that the user interacts with an ordinary natural language document rather than a relatively unfamiliar diagram.<P>The acronym means `What You See Is What You Meant'. The feedback text presented to the user (What You See) reveals the knowledge that has been encoded during the interaction so far (What You Meant). Documentation of knowledge bases becomes automatic, since the system is designed to produce a description in natural language of any knowledge base in any state of completion. The only limitation is that the knowledge base must conform to an ontology which from the user's point of view is fixed. If this ontology proves insufficient, it must be extended by a programmer; the user cannot add new concepts because the system would lack the linguistic resources to express them. 2000-09-18 http://www.itri.bton.ac.uk/projects/WYSIWYM/wysiwym.html knowledge editing WYSIWYM
PiantaTovena99, Pianesi-etal99 Pianta, Tovena, Pianesi 1997 1999 Italian XIG generates sentences from an interlingua content representation in the application area of tourist information. C-STAR II generator producing sentences from an interlingua content representation (Cstar Interchange Format): target of Cstar project is to build a speech-to-speech translation system able to deal with spontaneous speech in the application area of tourist information. 2000-06-03 http://ecate.itc.it:1024/projects/cstar/ tourist info XIG
Swartout81 Swartout 1981 1981 English XPLAIN 2000-06-03 XPLAIN
Reithinger87, Allgayer-etal89-xtra Reithinger, Allgayer, Harbusch, Kobsa 1986 1990 German XTRA Natural-Language Access System to Expert Systems incorporating generation functionalities Natural-Language Access System to Expert Systems incorporating generation functionalities 2000-06-20 XTRA
Stenzhorn2002-nlpxml, Stenzhorn2002-diplom Stenzhorn 2002 ongoing independent XtraGen Template-based generation with XML and Java interfaces XtraGen is a recently developed commercial software system for the flexible, real-time generation of natural language via sophisticated templates that can be easily integrated into real-world, industrial application environments through its open XML- and Java-based implementation and interfaces. A new library also allows to incorporate XtraGen in Java Server Pages directly with simple tags and no additional programming code. The actual generation kernel is founded in the fundamental ideas employed by TG/2 and other advanced template-based systems but incorporates a completely new and independent object-oriented design and implementation. 2003-09-11 template http://www.dfki.de/~holger/xtragen.html TG/2 XtraGen
Channarukul99, McRoy-etal2000, McRoy-etal99 McRoy, Channarukul, Ali 1999 2000 English YAG YAG is a real-time, general-purpose, template-based generation system YAG is a real-time, general-purpose, template-based generation system 2000-06-20 http://tigger.cs.uwm.edu/~nlkrrg YAG
Gabriel88 Gabriel 1988 1988 English YH simulation of conscious processes of writing simulation of conscious processes of writing 2000-01-01 simulation YH
VealeConway94 Veale, Conway 1994 1994 Sign Zardoz an English to Sign-Language Translation System an English to Sign-Language Translation System 2000-06-20 Zardoz
Jung-etal89 Jung, Kresse, Reithinger, Schäfer 1989 1989 German ZORA 2000-06-20 ZORA
BoyerLapalme85 Boyer, Lapalme 1985 1985 French generation of paraphrases generation of paraphrases 2000-01-01 BoyerLapalme
Carroll:1980-teaching Carroll 1980 1980 English CALL system for teaching English CALL system for teaching English 2000-01-01 teaching English Carroll
Cerbah92-coling, Cerbah92-ecai, Cerbah91, Cerbah92-avignon Cerbah 1991 1992 French Causal explanation generation Causal explanation generation 2001-11-22 Cerbah
Danlos87, Danlos87-kempen Danlos 1985 1988 French, English use of discourse grammar use of discourse grammar 2000-01-01 Danlos
deRoeckLowden86 de Roeck, Lowden 1985 1986 English generation of paraphrases generation of paraphrases 2000-06-03 deRoeck
Fournier91 Fournier 1991 1991 French A meaning-text text generator A meaning-text text generator 2001-11-22 MTM Fournier
Fraczak-etal98-subway, Lizogat2000, Fraczak98-phd, Fraczak-etal98-taln, FraczakLapalme99 Fraczak, Lapalme 1996 1998 French Generation of subway directions Generation of subway directions 2002-02-04 directions Fraczak, Lapalme
Friedman69 Friedman 1969 1969 English testbed for transformational grammar testbed for transformational grammar 2000-01-01 Friedman
GreenCarberry99-cl, GreenCarberry99, GreenCarberry95 Green, Carberry 1994 1994 English built to model the generation and interpretation of conversational implicatures The system generates indirect answers to Yes-No questions in English and was built to model the generation and interpretation of conversational implicatures 2000-06-20 conversational implicature, indirect answers, answers to Yes-No questions Green, Carberry
HakkaniOflazer98, Hakkani-etal96 Hakkani, Oflazer, Cicekli 1995 1998 Turkish Experimental generation Experimental generation focusing on the word order problem of 'free-word order' languages 2002-09-29 FUF Hakkani, Oflazer, Cicekli
 ??? Harper et al. 1969 1969 English random paragraph generation random paragraph generation 2000-01-01 Harper-etal
Herskovits73 Herskovitz 1973 1973 French semantic-based generation semantic-based generation 2002-02-04 Herskovitz
Hoffman94, Hoffman95 Hoffman 1994 1994 Turkish Experimental generation Experimental generation focusing on the word order problem of 'free-word order' languages 2002-09-29 Hoffman
Ishizaki1983, Ishizaki88 Ishizaki 1983 1988 Japanese generation of reports about terrorrist crimes generation of reports about terrorrist crimes 2000-01-01 news Ishizaki
Kalita89, KalitaShastri87, KalitaShende88 Kalita, Shastri, Shende 1987 1989 English natural language reports in an office environment natural language reports in an office environment 2000-06-20 business reports Kalita, Shastri, Shende
Klein65 Klein 1965 1965 English story generator story generator 2000-01-01 stories Klein-stories
KleinSimmons63 Klein 1963 1963 English coherent discourse generation coherent discourse generation 2000-06-03 Klein-coherence
Klein65-style Klein 1965 1965 English Multiple paraphrases of the same input texts Multiple paraphrases of the same input texts using the same set of dependency/Phrase-Structure rules with variant settings of rule probabilities for syntax and pronoun usage. 2000-06-03 paraphrase Klein-para
Klein-etal79, Klein-etal73 Klein 1973 1973 English program for generating murder mysteries on the basis of simulation data The output texts were the output of a simulation program for generating murder mysteries in a relational calculus semantic notation. The logic included some 1st order logical quantification strategies. The output of each time frame of the simulation was sent to a generator that used semantic/syntactic production rules. The outputs described the events in the simulations as as they were taking place. 2000-06-03 murder mysteries Klein-mysteries
Klein-etal74, Klein-etal77, Klein-etal76 Klein 1974 1977 English An exact model of V. Propp's Morphology of the Folktale An exact model of V. Propp's Morphology of the Folktale, formulated in and generated by the Meta-symbolic Simulation system. Program listing included, with comments that references to page numbers in Propp's book. Also contains myth texts derived from a model of the "Three Part Inventions" chapter of Levi-Strauss', THE RAW AND THE COOKED. <P><B>Comments from the author</B><BR> The English version of the Propp model appeared first as the UWCS Technical Report in 1974. The 1974 Tech Report version is exactly the same as the one that appeared in the 1977 book. <P>For historical reasons, the 1974 tech report is significant-- in December 1974, I mailed a copy to Claude Lévi-Strauss, whom I had never met. This led to his obtaining a visiting post for me in Paris at L'Ecole des hautes études en Science Sociale in 1976-- to work on modelling the thought processes he used in producing Mythologiques.<P> A revised French version of the paper, with a revised Propp model and an revised model for generating the 'double twist' transformation in the "Three Part Inventions" chapter of Levi-Strauss' The Raw & the Cooked was published in 1976. Parts of this French version are in English: a. the full set of rules for the Propp model are listed, with explanatory comments in English. b. the texts of the generated Russian tales are in English. 2000-09-26 stories Klein
Kreyss94-phd Kreyss 1994 1994 German content planning (PhD) content planning (PhD) 2000-06-20 Kreyss
Kukich:1987 Kukich 1987 1987 English connectionist-based sentence generation connectionist-based sentence generation 2000-01-01 Kukich
MeleroFont-Llitjos2001 Melero, Font-Llitjos 2000 ongoing Spanish Part of Microsoft's NLP effort Part of Microsoft's NLP effort 2001-06-01 PLNLP Melero, Font-Llitjos
Mellish:1986, MellishEvans89 Mellish, Evans 1987 1987 English text generation from plans text generation from plans 2000-01-01 Mellish, Evans
Memmi79 Memmi 1979 1979 French phrase generation (PhD) phrase generation (PhD) 2000-06-20 Memmi
Meunier97-phd Meunier 1997 1997 French TAG-based generation (PhD) TAG-based generation (PhD) 2000-06-20 TAG Meunier
Moghrabi:1980 Moghrabi 1980 1980 French generation of cooking recipes generation of cooking recipes 2000-01-01 recipes Moghrabi
Mooney93, Mooney-etal91 Mooney 1991 1993 English macro structure of explanation texts macro structure of explanation texts 2000-06-20 explanation Mooney
Namer89, Namer90 Namer 1990 1990 French, Italian generation of summaries of Verdi's operas generation of summaries of Verdi's operas 2000-01-01 music Namer
OhRudnicky2000 Oh, Rudnicky 2000 2000 English English Corpus-based stochastic generation for spoken dialog systems English Corpus-based stochastic generation for spoken dialog systems 2000-06-20 stochastic http://fife.speech.cs.cmu.edu/Communicator/papers/ Oh, Rudnicky
ACLShapiro, Shapiro82-ajcl Shapiro 1975 1982 English generation from semantic networks generation from semantic networks 2000-06-03 Shapiro
SimmonsSLOCUM Simmons, Slocum 1972 1972 English sentence generation by using an ATN sentence generation by using an ATN 2000-01-01 Simmons, Slocum
Simonin:1987 Simonin 1985 1987 French optimization, i.e. restructuring of plans in order to describe the economic situation of different countries optimization, i.e. restructuring of plans in order to describe the economic situation of different countries 2000-01-01 economics Simonin
Karkaletsis-etal98, Spyropoulos-etal96 Spyropoulos, Karkaletsis 1995 1999 Greek automatic generation of software messages automatic generation of software messages 2000-06-03 online help Spyropoulos, Karkaletsis
Tait85 Tait 1985 1985 English An English Generator for a Case-Labelled Dependency Representation An English Generator for a Case-Labelled Dependency Representation 2000-06-03 Tait
Wagner-etal99, Wagner-etal97-aime, Wagner-etal94, Wagner-etal95, Wagner-etal95-iccs, Rassinoux-etal98 Wagner, Solomon, Baud, Michel, Juge, Rector, Scherrer, Rassinoux 1994 1999 English, French Medical domain generation medical domain generation on the basis of the large-scale GALEN medical ontology, including both term and procedure generation 2001-09-18 medical Wagner-med
Wanner97-phd Wanner 1993 1997 German lexical resources for generation (PhD) lexical resources for generation (PhD) 2000-06-20 Wanner
Wilcock93-msc Wilcock 1993 1993 English generation of complex English modal and auxiliary verb groups are presented in Japanese to a Japanese user A demonstration protoype of an interactive generation system, in which inquiries concerning factors needed in the generation of complex English modal and auxiliary verb groups are presented in Japanese to a Japanese user. The necessary factors are used as an SFG network of choices, and the inquiries are based on felicity conditions for the appropriateness of the choices. The prototype system can be used for iterative revision of the generated sentences. This suggests an approach to knowledge-based post-editing. 2000-06-20 SFG http://www.ling.helsinki.fi/kit/2001s/ctl310gen/GW-MScThesis/thesis.html interactive MT Wilcock
Wong75 Wong 1975 1975 English sentence generation from a semantic network sentence generation from a semantic network 2000-01-01 Wong
Yang92, Yang98 Yang 1992 1998 Chinese systemic-functional oriented tactical generation of Chinese systemic-functional oriented tactical generation of Chinese 2000-06-20 Yang
Yngve61-random Yngve 1961 1961 English random generation with a generative grammar random generation with a generative grammar 2000-01-01 Yngve
Murray2002-mind Murray 2002 2002 English Mind-1.1 Concept-based generation system comprising a miniature Mind Tutorial AI in JavaScript for Microsoft Internet Explorer 2002-12-21 http://mind.sourceforge.net/jsaimind.html software tutorial Mind.Forth Mind
Lareau2002 Lareau 2002 2002 French Sentence Garden A generator of paraphrases based on Igor Mel'cuk's Meaning-Text Theory Sentence Garden is an experimental prototype generator of paraphrases based on Igor Mel'cuk's Meaning-Text Theory. The system is language-independent in theory, although only a fragment of description of a French lexicon and grammar has been developed for it so far. The lexicon it uses is the DiCo, a Meaning-Text dictionary that is developed at the U. of Montreal by Mel'cuk and Polguère. 2002-12-22 MTM http://www.fas.umontreal.ca/ling/olst/lareau paraphrase SenGarden
Habash00-oxygen, Habash2001, Habash-etal-2001, HabashDorr2001 Habash 1999 2002 independent OxyGen A language independent Linearization Engine A language independent Linearization Engine. OxyGen takes an input expression in the form of the Abstract Meaning Representation (AMR) developed for the <a href=Nitrogen.htm > Nitrogen </a> generator and works out a linear ordering for the specified lexical items. The output of the linearization process is a word lattice. A language (oxyL) is specified for writing linearization rules. The linearization process is hybrid, combining symbolic and statistical information. The system has been used in machine translation efforts such as <a href=http://hanne.umiacs.umd.edu:11111/chinmt/ >ChinMT</a> and others from Bonnie Dorr. 2002-12-22 Oxygen
Brun-etal2000-inlg, Dyetman-etal00-mda, BrunDymetman02-mda Xerox: DCM (Document Content Models) at XRCE (Xerox Research Center Europe) 2002 ongoing English, French, German MDA Multilingual Document Authoring system MDA (Multilingual Document Authoring) is an interactive natural language generation system which uses a unification grammar formalism for the specification of well-formedness conditions both on the semantics and on the surface realization of documents. The MDA project provides interactive tools, such as context-aware menus, for assisting monolingual writers in the production of multilingual documents. The author's choices have language-independent meanings (example: choosing between a solution and an emulsion in a drug description document), which are automatically rendered in any of the languages known to the system, along with their grammatical consequences on the surrounding text. Although the author is not explicitly following standards, the text produced by the system is implicitly controlled both:<UL><LI> Syntactically: the choice of the standard term for expressing a given notion is under system control, as is the choice between grammatical variants (such as active/passive sentences) for expressing a given information;</LI><LI>Semantically: the consequences of a choice somewhere are reflected across the whole document, the author cannot forget to provide some information that the system requires, dependencies between semantic parameters such as gender and pregnancy can be described.</LI></UL> 2003-01-19 http://www.xrce.xerox.com/competencies/content-analysis/dcm patient information leaflets MDA
Ceusters 1994 1996 English, Dutch, French Anthem Generation of medical diagnoses The aim of the ANTHEM (Advanced Natural Language Interface for Multilingual Text Generation) project was to have ready in September 1996 an API for medical systems able to generate out of medical diagnosis sublanguage statements in French or Dutch, language independent representations of the utterances that will generate translations of the original input in Dutch, French or German, and to perform automatic encoding following the ICD-10 classification rules. A second aim of ANTHEM was to use the same sublanguage approach for the analysis of standardised medical diagnosis expressions such as those used in international classification systems in order to facilitate cross-mapping to other systems and a smooth transition from older to new versions.<P> ANTHEM undertook testing of the final prototypes in a real world environment. Validation of the prototypes were carried out from a threefold perspective: linguistically (by testing the performance of the various subsystems with respect to automatic encoding and translation), technically (by assessing the flexibility with which the APIs can be integrated in existing healthcare applications) and operationally (by allowing end-users to comment on the overall user-friendliness and practical performance of the system). 2003-01-19 CAT-2 http://www.hltcentral.org/projects/detail.php?acronym=ANTHEM healthcare, medical diagnoses Anthem
Axelrod2000 Axelrod 2000 2000 English IBM Flight Information System IBM Flight Information System 2003-05-01 flight information IBMflight
Roh-etal01 Roh, Kang, Lee 2001 2001 Korean XEplainer Generation from a database for homeshopping sites Generation from a database for homeshopping sites 2003-05-01 homeshopping XEplainer
Striegnitz2000, GardentThater01 Striegnitz, Gardent 2000 2001 English InDiGen Integrated generation from declarative grammar-semantics-goal specifications The goal of this project is to develop an integrated approach to discourse and sentence planning which captures the interaction of discourse marker selection, ellipsis, and discourse structure. This project will focus on (a) empirical investigations and the formalization of the corresponding linguistic and non-linguistic constraints, and (b) the declarative modelling and implementation of those constraints by means of first order logic theorem proving and expressive concurrent constraint languages. <P> The system has an <B>online demo</B> at <A HREF=http://www.coli.uni-sb.de/cl/projects/indigen/demo/>here</A>. 2003-05-04 LTAG http://www.coli.uni-sb.de/cl/projects/indigen.html general SPUD InDiGen
Ratnaparkhi01 Ratnaparkhi 2001 2001 English Air travel reservation dialogue system Explores a surface generation framework that allows a dynamic word re-ordering in the context of a conversational system. The system attempts to duplicate the use of "word re-ordering" in order to maximize their communicative effectiveness. The surface NLG module expresses new information differently from old information. The input to the system consists of attribute-value pairs, some mandatory, some optional. Generation is a carried out by a hybrid statistical and grammar-based process. 2003-06-01 hybrid air travel Ratnaparkhi
WilksBy2003 Wilks, By, Lee, Ballim 2000 ongoing English ViewGen ViewGen (Viewpoint Generation) is an approach to dialogue understanding and belief modelling being implemented in a suite of programs. It includes natural language generation for hypertext as part of this overall project. ViewGen (Viewpoint Generation) is an approach to dialogue understanding and belief modelling implemented in a suite of programs. It includes natural language generation for hypertext and builds on a particular philosophical position concerning discourse and belief. <p> <strong>Extract from website: </strong> NLG systems often represent user-related information into a separate knowledge source, i.e., a user model. Typically, a user model will track some of user's beliefs, goals, interests, differences between a user's and system's beliefs, and nested beliefs (i.e., an agent's beliefs about other agent's beliefs). In our view, what is needed is an independent, powerful agent modelling component which can be used to support both understanding and generation. The advantage of using such a component is that potentially it could be reused between application domains and also between generators. Therefore we are now attempting to integrate ViewGen into a hypertext generation system. From that perspective, an important advantage of ViewGen is its account for common knowledge (i.e., default ascription) and stereotypes, as this will allow the modelling of different classes of users and tasks and the use of the most relevant one(s). In addition, the topic environments will restrict the search for relevant information to be included in the generated text. 2003-09-07 http://www.dcs.shef.ac.uk/nlp/viewgen/ hypertext ViewGen
DorrGaasterland02 Dorr, Gaasterland 2001 2002 English CONGEN ConGen (CONnective GENerator): translation of expressions involving tense and aspect ConGen (CONnective GENerator): translation of expressions involving tense and aspect. The system takes time-stamped dependency trees and derives appropriate temporal connectives. It is assumed to be embedded within a full generation system. 2003-10-29 MT ConGen
schnell02-umset-konzep-sprac Schnell 2001 2002 German Demosthenes Generates spoken monologues on different vine grapes (Riesling, Pinot noir, ...) and their respective wines. Main focus is on prosody generation. Generates spoken monologues on different vine grapes (Riesling, Pinot noir, ...) and their respective wines. Main focus is on prosody generation. The system is a concept-to-speech generation system containing stages adapted from Levelt's model of language production. Modules are present for all the major stages of language production and output containing fine-grained information about prosodic features is produced as final result. The system is implemented in Haskell and can produce output suitable for a variety of speech synthesis components. 2003-11-16 wine descriptions ex-Demosthenes
Gamon-etal02, Corston-Oliver-etal02, Smets-etal02 Gamon, Ringger, Corston-Oliver, Moore 2001 ongoing German, French, English Amalgam Amalgam is a sentence realization component based on data-driven, machine learning techniques. Amalgam accepts as input a logical form graph capturing the meaning of a sentence. Amalgam transforms the logical form into a fully articulated tree structure from which an output sentence is read. Amalgam is a novel system developed in the Natural Language Processing group at Microsoft Research for sentence realization during natural language generation. Sentence realization is the process of generating (“realizing”) a fluent sentence from a semantic representation. From the outset, the goal of the Amalgam project has been to build a sentence realization system in a data-driven fashion using machine learning techniques. Amalgam accepts as input a logical form graph capturing the meaning of a sentence. Amalgam transforms the logical form into a fully articulated tree structure from which an output sentence is read.To date, we have implemented Amalgam for both German and French, with English in the works.<H3>Example of Amalgam input</H3><IMG SRC=../images/amalgam-input.jpg><H3>Example of Amalgam structural output</H3><IMG SRC=../images/amalgam-output.jpg> 2003-11-16 machine learning http://research.microsoft.com/nlp/Projects/AmalgamProject.aspx technical manuals Amalgam
Piwek03-mnlg Piwek 2003 2003 English MNLG NMLG: Multimodal Natural Language Generation module NECA MNLG is a fully implemented Multimodal Natural Language Generation module. The MNLG is deployed as part of the NECA system which generates dialogues be-tween animated agents. The generation module supports the seamless integration of full grammar rules, templates and canned text. The generator takes input which allows for the specification of syntactic, semantic and pragmatic constraints on the output. <P>The input is processed in a pipeline that consists of the following modules in this order: <UL> <LI>A DIALOGUE PLANNER, which produces an abstract description of the dialogue (the dialogue plan). <LI>A MULTI-MODAL NATURAL LANGUAGE GENERA-TOR which specifies linguistic and non-linguistic real-izations for the dialogue acts in the dialogue plan. <LI>A SPEECH SYNTHESIS MODULE, which adds infor-mation for Speech. <LI>A GESTURE ASSIGNMENT MODULE, which controls the temporal coordination of gestures and speech. <UL> 2003-11-17 mixed animated interfaces MNLG
Dale-etal2005-routes, Dale-etal2003-coral, Dale-etal2002-coral, Geldof2003-enlgw, GeldofDale-2002 Dale, Geldof, Prost 2001 2003 English Coral Automatic generation of navigational route descriptions Route directions have been the subject of extensive research in various areas, including cognitive science, psychology, artificial intelligence and natural language generation. Our primary perspective is that of natural language generation. Thus our aim is to investigate in how far natural language generation technology can improve the efficiency of automatically generated route descriptions. The current technological context allows for information delivery in the physical context of usage via mobile devices. This is particularly relevant for instructional discourse, such as route descriptions. 2008-04-12 http://www.ics.mq.edu.au/~coral/index.html route descriptions Coral
KranzdorfGrefahn92-itex Kranzdorf, Grefahn 1992 1992 German ITEX A hybrid approach to text planning A hybrid approach to text planning that investigates the flexible relationship between macrostructure and microstructure. Surface generation is done with an LFG-based generator. 2004-12-18 hybrid medical LFG-f ITEX
DoerreMomma87-gwai, MommaDoerre87 Dörre, Momma 1987 1987 German Generation from LFG f-structures A method for generating sentences from LFG f-structures 2004-12-18 LFG general LFG-f
reitter04-mug, Reitter-etal2004-ui Reitter 2003 ongoing independent MUG Multimodal functional unification generation When grammar-based techniques for natural language generation (and analysis alike) find their way into collaborative projects or actual application, big grammars tend to become hard to extend and debug. The MUG system represents a new tool set with a graphical debugging environment for functional unification grammars, which is designed to help grammar developers inspect the results of their work. The particular formalism supported is Multimodal Functional Unification Grammar, which is similar to Functional Unification Grammars, but supports several coordinated modes, such as voice prompts or structural and/or language-based screen displays. For each input description, the grammar can generate a range of coherent realization variants, which are ranked by a scoring function in order to optimize the output towards situational and device-related factors. 2004-12-18 FUG http://www.reitter-it-media.de/compling/mug/index.html general MUG
Benamara-2004-inlg Benamara 2004 2004 English, French WEBCOOP Logic-based model for question answering A logic-based model for the accurate generation of intensional responses within a co-operative question-answering framework. Generation is template-based. 2004-12-19 template transportation WEBCOOP
Carroll-etal2000 Carroll, Nicolov, Smets, Shaumyan, Weir 2000 2003 English LEXSYS A wide-coverage lexicalized grammar A wide-coverage lexicalized grammar 2004-12-19 TAG general LEXSYS
Stede2002-coling, ChiarcosStede-2004-inlg Stede, Chiarcos 2003 ongoing English, German Polibox Hypertext generator Hypertext generator 2004-12-19 Polibox
Habash-2004-inlg Habash 2004 2004 English EXERGE EXpansivE Rich Generation for English EXpansivE Rich Generation for English using statistical methods and an N-gram language model 2004-12-19 n-grams Exerge
MarciniakStrube-2004-inlg Marciniak, Strube 2003 2004 English Generator for route descriptions Generation of route descriptions by apply automatic learning techniques to drive a TAG-based generation system. The training corpus is for route descriptions. 2004-12-19 TAG route descriptions Marciniak
PaivaEvans-2004-inlg Paiva, Evans 2004 2004 English Stylistic control for generation System that derives stylistic control for generation from a corpus analysis of a range of language varieties that will be covered. The corpus analysis follows factorial analysis (Biber) and applies a correlation technique to ensure that generated texts are situated appropriately within the space of possibilities. 2004-12-19 LEXSYS, RICHES medical Paiva-style
PanShaw-2004-inlg Pan, Shaw 2004 2004 English SEGUE Hybrid case-based surface generation Hybrid case-based surface generation 2004-12-19 general SEGUE
Foster-2004-inlg, White-2004-inlg Foster, White, Moore 2004 2004 English COMIC COnversational Multimodal Interaction with Computers Multimodal dialogue system that adds a spoken-language dialogue interface to a CAD-like application used in bathroom sales situations to help clients redesign their rooms. Input to the system includes speech, handwriting and pen gestures; the output combines synthesized speech, a 'talking head' avatar and control of teh underlying application. 2004-12-19 CCG http://www.hcrc.ed.ac.uk/comic/ interior design COMIC
Klarner-2004-inlg Klarner 2004 2004 English HYPERBUG A hybrid NLG system for generating dialogue: 'Hybrid pragmatically embedded realization with bottom-up generation' Mixes shallow and deep generation for dialogue generation. Currently used in three domains: home audio-visual management, model train control, and B2B e-procurement. 2004-12-19 HYPERBUG
Claassens2005-ola Claassens 2004 2005 English OLA English Question/Answer (Q&A) system, template-based and not very sophisticated, but the open source NLG system can be used in many practical and online cases. This English Question/Answer (QA) system is template-based and is not very sophisticated. However, this open source NLG system can be used in many practical and online cases. <P>Till now, this QA system can be used to answer questions that are related to rids. This domain has a very restricted lexicon. Therefore, it is an easy domain in which such a system can be applied to. <P>O.L.A. is developed in Linux environment in AWK programming code (see website). The gaps of the randomly chosen template sentences are filled with usage of external data. Furthermore, an example of an avatar will be generated (form: emoticon). 2005-03-03 http://pi0959.uvt.nl/s883247/acklin.html OLA
Carenini2001-gea, CareniniMoore2000 Carenini 2000 2002 English GEA Generator of Evaluative Arguments (GEA): GEA is a fully-implemented, complete and modular NLG system for generating user tailored evaluative arguments. GEA is a fully-implemented, complete and modular NLG system for generating user tailored evaluative arguments. GEA covers all aspects of generating evaluative arguments from selecting and organizing the content of the argument, to expressing the selected content into natural language. For content selection and organization, GEA applies an argumentation strategy based on guidelines from argumentation theory. For expressing the content into natural language, GEA relies on a set of techniques that extend and integrate previous work in computational linguistics on micro-planning and realizing evaluative arguments. Finally, a quantitative model of user preferences expressed as an additive multiattribute value function (AMVF) is the key knowledge source used by GEA in tailoring the content, organization and phrasing of the generated arguments to its users. 2005-05-21 http://www.cs.ubc.ca/%7Ecarenini/THESIS/gea.html arguments GEA
AbellaKender99 Abella, Kender 1999 1999 English Spatial language generation Generating sentences from images employing spatial relations 2005-09-15 scene descriptions Abella
Boyd98 Boyd 1998 1998 English TREND Time-series description Time-series description 2005-09-15 time-series TREND
Karamanis-etal2004, Karamanis2003, Karamanis-etal-2004-inlg Karamanis 2002 2004 English SEEC System for Evaluating Entity Coherence A text structuring module for NLG that implements a search-oriented, corpus-based methodology for evaluating metrics of entity coherence based on different vers ions of Centering Theory.Various permutations of input expressions are evaluated with respect to the corpus-derived metrics. 2005-09-27 text structuring SEEC
CziferszkyWinter2002 Cziferszky, Winter 2002 2002 German Route description generation Route description generation 2005-09-27 route descriptions CziferszkyWinter
WilliamsReiter2005-enlg, WilliamsReiter2005-ijcai, WilliamsReiter2005-selrules, WilliamsReiter2004-errors, Reiter-etal2005-sgai Williams, Reiter 2003 2005 English SkillSum Planning content, structure, discourse-level and lexical choices for people with poor literacy and numeracy Planning content, structure, discourse-level and lexical choices for people with poor literacy and numeracy 2006-03-09 http://www.csd.abdn.ac.uk/~swilliam/skillsum/ Personalised basic skills summary reports SkillSum
Szilas2007 Szilas 2007 2007 English IDtension System implementing many aspects of narrative theory in order to support interactive drama System implementing many aspects of narrative theory in order to support interactive drama. The narrative planning and realisation is intended to be medium-independent; early version of the system produces text descriptions by template generation. 2007-09-15 templates narrative IDtension
Levison-etal2001, LessardLevison95, LevisonLessard96, LevisonLessard92 Lessard, Levison 1986 ongoing English, French VINCI VINCI is a natural language generation environment used since 1986 for language learning and testing, subject testing, and modelling of diverse linguistic phenomena ranging from word-formation, to narrative structure, to generation of verbal humour. <P>VINCI is a natural language generation environment under development for the past 20-odd years. It has been used since 1986 for language learning and testing, subject testing, and modelling of diverse linguistic phenomena ranging from word-formation, to narrative structure, to generation of verbal humour. Most research has been done in French or English, but the system was designed to be multilingual and some work has been done in other languages as well.<P>The system includes a variety of features, including:<UL><LI>a semantic metalanguage which is not language-specific</LI><LI>lexical preselection mechanisms which allow initial choice of lexical items in some natural language and inheritance of traits from these items in subsequent syntactic development</LI><LI>a system of attributes which provides the 'glue' between semantics, syntax, morphology and lexicon; attributes may be partially ordered, compounded, and deconstructed</LI><LI>a syntactic mechanism based on a context-free grammar with attribute attachment, which allows inheritance, guarded syntax rules allowing for conditional development of child nodes, and transformations of syntax trees</LI><LI>lexical representation based on syntactic and semantic attributes, morphological traits, orthography, phonology, and frequency; also included are lexical pointer mechanisms to permit representation of relations such as hyponymy, synonymy, etc.</LI><LI>morphology rules based on attribute values, orthography, phonology, and preceding and following context</LI><LI>lexical transformation devices which permit the synthesis by rule of sets of derived or compound words, either statically or dynamically</LI><LI>error analysis mechanisms which take some user input and compare it to some set of rule-governed candidate parses at the phonological, phonographic, orthographic, morphological, lexical, syntactic and semantic levels.</LI></UL><P>VINCI is embedded in an editing environment (IVI) which provides for easy modification of language specifications, including template-based editing of lexical records. IVI also permits the use of driver programs written in C or some other language. The program is written in C and runs under Solaris, Linux, and Windows (using Cygwin). It is freely available from the project website, which also contains documentation on the system, sample language descriptions, and references to the papers and articles to which the project has given rise. 2007-09-15 http://www.cs.queensu.ca/CompLing/ CALL VINCI
CheongYoung2006 Cheong, Young 2005 2006 English Suspenser A framework for narrative generation Suspenser, a computational model of narrative generation that takes as input a given story world and constructs a narrative structure intended to evoke the desirable level of suspense from the reader at a specific moment in the story. 2008-01-12 narrative Suspenser
KamalMellish-2004-inlg Kamal, Mellish 2004 2004 English STAGE Generating using an Assumption-Based Truth Maintenance System (ATMS) The STylistics-Aware GEnerator (STAGE) system brings together systemic-functional linguistic resources and established an AI search mechanism. The generator uses systemic networks tranlsated into a particular form in order to generate nature language texts that can build on the ATMS in order to incorporate surface constraints. 2008-01-16 SFG general STAGE
Portet-etal2007, Reiter2007-data-to-text Portet, Reiter, Hunter, Sripada 2007 ongoing English BabyTalk BabyTalk is developing computerised tools to summarise and interpret data in the Neonatal Intensive Care Unit (NICU). BabyTalk is investigating ways of summarising and presenting patient information to medical professionals and family members. Our focus is on data in the Neonatal Intensive Care Unit. 2008-04-18 http://www.csd.abdn.ac.uk/research/babytalk medical BabyTalk
Reiter-etal2005-words, Sripada-etal2003-mousam Reiter, Yu, Hunter, Sripada, Davy 2005 ongoing English SumTime-Mousam Weather Forecast System Data-to-Text architecture for marine weather reports 2008-04-18 weather SumTime SumTime-Mousam
Portet-etal2007, Reiter2007-data-to-text Portet, Reiter, Hunter, Sripada 2005 2007 English SumTime-Neonate Medical Care Reports Data-to-Text architecture for medical reports: see also BabyTalk 2008-04-18 medicine SumTime SumTime-Neonate
Yu-etal2005 Reiter, Yu, Hunter, Mellish 2005 2005 English SumTime-Turbine System for generating textual summaries of sensor data from a gas turbine Data-to-Text architecture for sensor data from a gas turbine 2008-04-18 industrial SumTime SumTime-Turbine
Gupta-etal2007, Gupta-etal2007-ENLG, Gupta-etal2008 Gupta, Walker, Romano 2007 2008 English POLLy POliteness in Language Learning The system contains a generalised mechanism for generating utterances and dialogues, drawing on a sociolinguistics theory of politeness. It combines a Spoken Language Generator (using full NLG techniques) with an AI planner to model linguistic politeness in collaborative task oriented dialogue with the ultimate goal of providing a stimulating environment and a fun way for learning politeness in English as a Second Language (ESL). <P>Dialogues are demonstrated within Embodied Conversational Agents (ECA) capable of dialogues with human ESL learners, interacting with them using spoken language to practice politeness in role play situations.<P><IMG src=../images/polly.jpg> 2008-06-14 ESL Polly
MariesseWalker2007, MariesseWalker2008 Mairesse, Walker 2005 2008 English PERSONAGE Personage is a full natural language generator that can convey various personality traits. Personage is a full natural language generator that can convey various personality traits by controlling parameters derived from the psychology literature. Personage-PE---an extension of Personage---uses statistical models trained on human judgements to predict the optimal parameters given target personality scores along the Big Five dimensions of personality. <p>A Java real-time demo is available at <a href=http://mi.eng.cam.ac.uk/~farm2/personage/personage.html>http://mi.eng.cam.ac.uk/~farm2/personage/personage.html</a> 2008-04-18 http://mi.eng.cam.ac.uk/~farm2/personage/personage.html Affect PERSONAGE
AllmanBeale2006 Allman, Beale, Denton 1993 2008 English, Korean, Jula, Kewa TBTA The Bible Translator's Assistant The Bible Translator's Assistant is a multilingual natural language generator based on linguistic universals and typologies. This project is intended to generate rough drafts of the Bible and many community development articles in the world's 3000+ minority languages. Experiments indicate that these rough drafts quadruple the productivity of experienced mother tongue translators. (<a href=../images/tbta-allman2008.pdf>Cached unpublished paper</a>) 2008-06-14 UG http://www.thebibletranslatorsassistant.com/ translation TBTA
Dalianis99-volvex Dalianis 1996 1999 English VOLVEX Validation Of Specifications by Natural Language Generation for VOLVO expressed in STEP/EXPRESS STEP Application Protocols (APs) are often very large and complicated general descriptions of different domains mainly within the manufacturing industry. STEP AP's are expressed in the EXPRESS language which is a static modeling language of Entity-Relationship type. Users of STEP AP have often large problems to understand the whole AP and therefore they need a tool which helps them to validate the STEP AP. The tool we are proposing is a natural language English paraphraser of the STEP AP. The VOLVEX system supports automatically building a domain lexicon from one STEP AP and with this lexicon automatically translate one arbitrary STEP AP EXPRESS file into a Prolog based format to be used by a Natural Language Generator called ASTROGEN. The translation from EXPRESS format to EXPRESS Prolog format and the construction of the domain lexicon is carried out by a set of Perl programs. 2008-12-10 http://people.dsv.su.se/~hercules/papers/volvexhandbook.html formal specifications ASTROGEN VOLVEX
Turner-etal2008, Turner-etal2007 Turner, Sripada, Reiter, Davy 2006 2009 English RoadSafe Automatically Generating Advisory Text for Road Maintenance Vehicle Routing A project for "Automatically Generating Advisory Text for Road Maintenance Vehicle Routing". The RoadSafe project is intended to<UL><LI>use Knowledge Aquisition techniques to understand how humans write textual instructions for road maintenance vehicle routing.</LI><LI>produce a system capable of automatically evaluating a region's geographical data combined with the weather forecast for 10000s of points in that region to provide textual routing and de-icer spread rate instructions</LI><LI>utilise Aerospace and Marine International's expert forecasters in order to post-edit generated advisory texts and therefore improve the performance of the system. </LI></UL> 2009-01-24 http://www.csd.abdn.ac.uk/~rturner/RoadSafe/ geodata RoadSafe
Belz2007 Belz 2006 2009 English pCRU Probabilistic Generation of Weather Reports Probabilistic Generation of Weather Reports 2009-01-24 statistical weather pCRU
GalanisAndroutsopoulos2007 Galanis, Androutsopoulos 2006 2008 English, Greek NaturalOWL Description of concepts defined in OWL ontologies 2009-01-24 http://pages.cs.aueb.gr/nlp/software.html NaturalOWL
Reiter2007-data-to-text Reiter 2007 2009 General SimpleNLG A simple Java-based generation framework A simple Java-based generation framework. SimpleNLG is a relatively simple Natural Language Generation realiser, with a Java API. It has less grammatical coverage than many other realisers, but it does not require in-depth knowledge of a syntactic theory to use. The core of the package is the realiser, lexicon, and features packages 2009-01-24 http://www.csd.abdn.ac.uk/~ereiter/simplenlg/ general SimpleNLG
ThomasSripada2007 Thomas, Sripada, Reiter 2006 2009 English Atlas.txt Making geo-referenced data available via natural language generation The Atlas.txt project aims to produce textual summaries of geo-referenced data which can be read out via a screenreader. Such data are often communicated via maps are so are inaccessible to the visually impaired population. 2009-01-24 http://www.csd.abdn.ac.uk/~ssripada/atlas/ geodata SimpleNLG Atlastxt