aller au contenu
Grammatical trigrams : a probabilistic model of link grammar Aperçu de cet ouvrage
FermerAperçu de cet ouvrage
Vérifiant…

Grammatical trigrams : a probabilistic model of link grammar

Auteur : John D Lafferty; Daniel D Sleator; Davy Temperley
Éditeur : Pittsburgh, Pa. : School of Computer Science, Carnegie Mellon University, [1992]
Collection : Research paper (Carnegie Mellon University. School of Computer Science), CMU-CS-92-181.
Édition/format :   Livre : AnglaisVoir toutes les éditions et les formats
Base de données :WorldCat
Résumé :
Abstract: "In this paper we present a new class of language models. This class derives from link grammar, a context-free formalism for the description of natural language. We describe an algorithm for determining maximum-likelihood estimates of the parameters of these models. The language models which we present differ from previous models based on stochastic context-free grammars in that they are highly lexical. In  Lire la suite...
Évaluation :

(pas encore évalué) 0 avec des critiques - Soyez le premier.

Sujets
Plus comme ceci

 

Trouver un exemplaire dans la bibliothèque

&AllPage.SpinnerRetrieving; Recherche de bibliothèques qui possèdent cet ouvrage...

Détails

Format : Livre
Tous les auteurs / collaborateurs : John D Lafferty; Daniel D Sleator; Davy Temperley
Numéro OCLC : 26847322
Notes : "To appear in Proc. of the 1992 AAAI Fall Symp. on Probabilistic Approaches to Natural Language."
"September 1992."
Description : 10 pages ; 28 cm.
Titre de collection : Research paper (Carnegie Mellon University. School of Computer Science), CMU-CS-92-181.
Responsabilité : John Lafferty, Daniel Sleator, Davy Temperley.

Résumé :

Abstract: "In this paper we present a new class of language models. This class derives from link grammar, a context-free formalism for the description of natural language. We describe an algorithm for determining maximum-likelihood estimates of the parameters of these models. The language models which we present differ from previous models based on stochastic context-free grammars in that they are highly lexical. In particular, they include the familiar n-gram models as a natural subclass. The motivation for considering this class is to estimate the contribution which grammar can make to reducing the relative entropy of natural language."

Critiques

Critiques d’utilisateurs
Récupération des critiques de GoodReads...
Récuperation des critiques DOGObooks…

Tags

Soyez le premier.

Ouvrages semblables

Confirmez cette demande

Vous avez peut-être déjà demandé cet ouvrage. Veuillez sélectionner OK si vous voulez poursuivre avec cette demande quand même.

Données liées


<http://www.worldcat.org/oclc/26847322>
library:oclcnum"26847322"
library:placeOfPublication
library:placeOfPublication
rdf:typeschema:Book
schema:about
schema:about
<http://id.worldcat.org/fast/1034365>
rdf:typeschema:Intangible
schema:name"Natural language processing (Computer science)"@en
schema:contributor
schema:contributor
schema:creator
schema:datePublished"1992"
schema:description"Abstract: "In this paper we present a new class of language models. This class derives from link grammar, a context-free formalism for the description of natural language. We describe an algorithm for determining maximum-likelihood estimates of the parameters of these models. The language models which we present differ from previous models based on stochastic context-free grammars in that they are highly lexical. In particular, they include the familiar n-gram models as a natural subclass. The motivation for considering this class is to estimate the contribution which grammar can make to reducing the relative entropy of natural language.""@en
schema:exampleOfWork<http://worldcat.org/entity/work/id/29987363>
schema:inLanguage"en"
schema:isPartOf
schema:name"Grammatical trigrams : a probabilistic model of link grammar"@en
schema:publication
schema:publisher
wdrs:describedby

Content-negotiable representations

Fermer la fenêtre

Veuillez vous identifier dans WorldCat 

Vous n’avez pas de compte? Vous pouvez facilement créer un compte gratuit.