Korpus / Textsammlung
deu Subkorpus der TüBa-D/Z mit Diskursstruktur und Diskursrelationen
eng Corpus-based Semantic Composition Models for Phrases
deu Korpusbasierte Semantische Kompositionsmodelle für Phrasen
eng This subcorpus of the treebank TüBa-D/Z contains discourse structures and discourse relations. For the study on automatic classification, we use instances of two German temporal connectives that can also carry a non-temporal discourse relation, namely `während´ and `nachdem. Our data set – the connective occurrences from the current extent of the TüBa-D/Z, totaling about 60 000 senteces – contains 294 instances of nachdem and 527 instances of während. Where available, we used the syntactic annotation from the treebank; in the remaining cases, we used a syntactic parser (Versley and Rehbein, 2009) to provide syntax trees for the feature extraction.
2 Ressource(n)
restricted for individual
049dc086-b8c9-4bb6-be8e-1f68d83fd6bd
65d11ea3-b1b4-451c-b04f-a7162a220608
vorhanden
Zugang via Browser, PID zeigt auf Landing Page der Resource
Textsammlung
data-steward@semsprach.uni-tuebingen.de
corpus
Syntactisches corpus
geschrieben