The GF Resource Grammar Library Version 1.0 Author: Aarne Ranta Last update: %%date(%c) % NOTE: this is a txt2tags file. % Create an html file from this file using: % txt2tags --toc clt2006.txt %!target:html %!postproc(html): #NEW #NEW ==Plan== Purpose Background Coverage Structure How to use How to implement a new language How to extend the API #NEW ==Purpose== ===Library for applications=== High-level access to grammatical rules E.g. //You have k new messages// rendered in ten languages //X// ``` render X (Have (You (Number (k (New Message))))) ``` Usability for different purposes - translation systems - software localization - dialogue systems - language teaching #NEW ===Grammar as parser=== Often in NLP, a grammar is just high-level code for a parser. But writing a grammar can be inadequate for parsing: - too much manual work - too inefficient - not robust - too ambiguous Moreover, a grammar fine-tuned for parsing may not be reusable - for generation - for specialized grammars - as library #NEW ===Grammar as language definition=== Linguistic ontology: **abstract syntax** E.g. adjectival modification ``` AdjCN : AP -> CN -> CN ; ``` Rendering in different languages: **concrete syntax** Resource grammars have generation perspective, rather than parsing - abstract syntax serves as a key to expressions in different languages #NEW ===Usability by non-linguists=== Division of labour: resource grammars hide linguistic details Presentation: "school grammar" concepts, dictionary-like conventions API = Application Programmer's Interface Documentation: ``gfdoc`` IDE = Interactive Development Environment (forthcoming) Example-based grammar writing ``` render Ita (parse Eng "you have k messages") ``` #NEW ===Scientific interest=== Linguistics - definition of linguistic ontology - coping with different problems in different languages - sharing concrete-syntax code between languages - creating a resource for other NLP applications Computer science - datastructures for grammar rules - type systems for grammars - algorithms: parsing, generation, grammar compilation - domain-specific programming language (GF) - module system #NEW ==Background== ===History=== 2002: v. 0.2 - English, French, German, Swedish 2003: v. 0.6 - module system - added Finnish, Italian, Russian - used in KeY 2005: v. 0.9 - tenses - added Danish, Norwegian, Spanish; no German - used in WebALT 2006: v. 1.0 - approximate CLE coverage - reorganized module system and implementation - not yet (4/3/2006) for Danish and Russian #NEW ===Authors=== Janna Khegai (Russian modules, forthcoming), Bjorn Bringert (many Swadesh lexica), Carlos Gonzalia (Spanish cardinals), Partik Jansson (Swedish cardinals), Aarne Ranta. We are grateful for contributions and comments to several other people who have used this and the previous versions of the resource library, including Ana Bove, David Burke, Lauri Carlson, Gloria Casanellas, Karin Cavallin, Hans-Joachim Daniels, Kristofer Johannisson, Anni Laine, Wanjiku Ng'ang'a, Jordi Saludes. #NEW ===Related work=== CLE (Core Language Engine, [Book 1992 http://mitpress.mit.edu/catalog/item/default.asp?tid=7739&ttype=2]) - English, Swedish, French, Danish - uses Definita Clause Grammars, implementation in Prolog - coverage for SACTI corpus, [Spoken Language Translator (2001) http://www.cambridge.org/uk/catalogue/catalogue.asp?isbn=0521770777] - grammar specialization via explanation-based learning [LinGO Grammar Matrix http://www.delph-in.net/matrix/] - English, German, Japanese, Spanish, ... - uses HPSG, implementation in LKB - a check list for parallel grammar implementations [Pargram http://www2.parc.com/istl/groups/nltt/pargram/] - Aimed: Arabic, Chinese, English, French, German, Hungarian, Japanese, Malagasy, Norwegian, Turkish, Urdu, Vietnamese, and Welsh - uses LFG - one set of big grammars, transfer rules Rosetta Machine Translation ([Book 1994 http://citeseer.ist.psu.edu/181924.html]) - Dutch, English, French - uses M-grammars, compositional translation inspired by Montague - compositional transfer rules #NEW ==Coverage== ===Languages==== The current GF Resource Project covers ten languages: - ``Dan``ish - ``Eng``lish - ``Fin``nish - ``Fre``nch - ``Ger``man - ``Ita``lian - ``Nor``wegian (bokmål) - ``Rus``sian - ``Spa``nish - ``Swe``dish In addition, parts (morphology) of Arabic, Estonian, Latin, and Urdu API 1.0 not yet implemented for Danish and Russian #NEW ===Morphology==== Complete inflection engine - all word classes - all forms - all inflectional paradigms High-level access via ``ParadigmsX``; e.g. Swedish: - worst-case functions ``` mkV : (supa,super,sup,söp,supit,supen : Str) -> V ; ``` - common patterns ``` regV : (talar : Str) -> V ; irregV : (dricka, drack, druckit : Str) -> V ; ``` - irregular words in ``IrregX``: ``` draga_V : V = mkV (variants { "dra"; "draga"}) (variants { "drar" ; "drager"}) (variants { "dra" ; "drag" }) "drog" "dragit" "dragen" ; ``` #NEW ===Syntactic structures=== [Lang.png] #NEW ===Quantitative measures=== 67 categories 150 abstract syntax combination rules 100 structural words 350 content words in a test lexicon Lines of source code (4/3/2006): ``` abstract 1131 english 2344 german 2386 finnish 3396 norwegian 1257 swedish 1465 scandinavian 1023 french 3246 -- Besch + Irreg + Morpho 2111 italian 7797 -- Besch 6512 spanish 7120 -- Besch 5877 romance 1066 ``` #NEW ==Structure== #NEW ===Language-independent ground API=== #NEW ===Language-dependent paradigm modules=== #NEW ===Language-dependent syntax extensions=== #NEW ===Special-purpose APIs=== #NEW ===How to use as top-level grammar=== #NEW ===Parsing=== #NEW ===Treebank generation=== #NEW ===Treebank-based parsing=== #NEW ===Morphology=== #NEW #NEW ===Syntax editing=== #NEW ===Efficient parsing via application grammar=== #NEW ==How to use as library== ===Specialization through parametrized modules=== #NEW ===Compile-time transfer=== #NEW ===A natural division into modules=== #NEW ===Example-based grammar writing=== #NEW ==How to implement a new language== ===Ordinary modules=== #NEW ===Parametrized modules=== #NEW ===The kernel of the API=== #NEW ===How to proceed=== #NEW ==How to extend the API== #NEW ===Extend old modules or add a new one?===