DE eng

Search in the Catalogues and Directories

Hits 1 – 18 of 18

1
Inducing Language-Agnostic Multilingual Representations ...
BASE
Show details
2
Zero-Shot Cross-Lingual Transfer with Meta Learning ...
BASE
Show details
3
SIGTYP 2020 Shared Task: Prediction of Typological Features ...
BASE
Show details
4
Uncovering Probabilistic Implications in Typological Knowledge Bases ...
BASE
Show details
5
Back to the Future -- Sequential Alignment of Text Representations ...
BASE
Show details
6
What do Language Representations Really Represent? ...
BASE
Show details
7
What Do Language Representations Really Represent?
In: Bjerva, Johannes; Östling, Robert; Veiga, Maria Han; Tiedemann, Jörg; Augenstein, Isabelle (2019). What Do Language Representations Really Represent? Computational Linguistics, 45(2):381-389. (2019)
BASE
Show details
8
Copenhagen at CoNLL--SIGMORPHON 2018: Multilingual Inflection in Context with Explicit Morphosyntactic Decoding ...
BASE
Show details
9
Parameter sharing between dependency parsers for related languages ...
BASE
Show details
10
Multitask and Multilingual Modelling for Lexical Analysis ...
Bjerva, Johannes. - : arXiv, 2018
BASE
Show details
11
From Phonology to Syntax: Unsupervised Linguistic Typology at Different Levels with Language Embeddings ...
BASE
Show details
12
The Parallel Meaning Bank: Towards a Multilingual Corpus of Translations Annotated with Compositional Meaning Representations
In: 15th Conference of the European Chapter of the Association for Computational Linguistics ; https://hal.archives-ouvertes.fr/hal-01630960 ; 15th Conference of the European Chapter of the Association for Computational Linguistics, Apr 2017, Valencia, Spain. pp.242 - 247 (2017)
BASE
Show details
13
Tracking Typological Traits of Uralic Languages in Distributed Language Representations ...
BASE
Show details
14
Articulation rate in Swedish child-directed speech increases as a function of the age of the child even when surprisal is controlled for ...
BASE
Show details
15
One Model to Rule them all: Multitask and Multilingual Modelling for Lexical Analysis ...
Bjerva, Johannes. - : arXiv, 2017
BASE
Show details
16
The Parallel Meaning Bank: Towards a Multilingual Corpus of Translations Annotated with Compositional Meaning Representations ...
BASE
Show details
17
Rethinking intertextuality through a word-space and social network approach – the case of Cassiodorus
In: https://hal.archives-ouvertes.fr/hal-01279833 ; 2016 (2016)
Abstract: Continuous space representations of words are currently at the core of many state-of-the-art approaches in computational linguistics. The distributional hypothesis, summarised as: 'You shall know a word by the company it keeps' [Firth, 1957] is the basis of many such methods. In this paper we use this type of representation, which has seen little to no use in digital humanities, to rethink the concept of intertextuality. We present and use an alternative conceptual concept of intertextuality to ascertain how different persons are portrayed in a late antique letter collection, the Variae of Cassiodorus (ca. 485–585 AD). We combine this approach with the well-explored method of network analysis. 'The study of intertextuality is the study of a certain kind of relation between texts: One text quotes another or others.' [Edmunds, 2001]. Until recently, intertextuality has been pictured as an interaction between different texts, which has been restricted to the surface forms. We want to transcend this rather limited, one-dimensional concept of intertextuality by using high-dimensional word representations which effectively abstract away from such surface forms. Instead of conceptualising, e.g. Vergil, as the sum of his transmitted oeuvre, we represent him both as a node in a network, and a vector in high-dimensional space. In this way we overcome the border between text and historical person; a border which impedes the ascertaining of the intertextual impact of authors which are partially or not at all preserved. We create word-space representations based on the letters in the Variae, using methods based on distributional semantics [Mikolov et al., 2013a, Levy et al., 2015]. In antiquity, the editing and publication of letter collections was a fundamental tool for literary and cultural self-representation. Late antiquity witnessed the zenith of this practice with the publication of several such collections, both in Latin and in Greek. The Variae of Cassiodorus are an excellent example of this practice of self-representation [Bjornlie, 2012, Gillett, 1998]. In this paper, we will represent Cassiodorus, his contemporaries, and influential authors of the literary canon, such as Vergil, in one and the same network. This form of visualisation can generate a more nuanced view on how Cassiodorus constructs a cultural profile for himself and his peers. Indeed, the letters of Cassiodorus act as a meeting ground in which both the contemporaries of Cassiodorus, as well as the authors who shaped the intellectual outlook of Cassiodorus and his peers, interact with each other.
Keyword: [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]; [SHS.HIST]Humanities and Social Sciences/History; cassiodorus; intertextuality; late antiquity; social networks; word embeddings
URL: https://hal.archives-ouvertes.fr/hal-01279833/file/bjerva_praet_2016_prereview.pdf
https://hal.archives-ouvertes.fr/hal-01279833
https://hal.archives-ouvertes.fr/hal-01279833/document
BASE
Hide details
18
Semantic Tagging with Deep Residual Networks ...
BASE
Show details

Catalogues
0
0
0
0
0
0
0
Bibliographies
0
0
0
0
0
0
0
0
0
Linked Open Data catalogues
0
Online resources
0
0
0
0
Open access documents
18
0
0
0
0
© 2013 - 2024 Lin|gu|is|tik | Imprint | Privacy Policy | Datenschutzeinstellungen ändern