Spelling suggestions: "subject:"very classification"" "subject:"vers classification""
1 |
Hybrid Methods for Acquisition of Lexical Information: the Case for VerbsJianguo, Li January 2008 (has links)
No description available.
|
2 |
Automatic induction of verb classes using clusteringSun, Lin January 2013 (has links)
Verb classifications have attracted a great deal of interest in both linguistics and natural language processing (NLP). They have proved useful for important tasks and applications, including e.g. computational lexicography, parsing, word sense disambiguation, semantic role labelling, information extraction, question-answering, and machine translation (Swier and Stevenson, 2004; Dang, 2004; Shi and Mihalcea, 2005; Kipper et al., 2008; Zapirain et al., 2008; Rios et al., 2011). Particularly useful are classes which capture generalizations about a range of linguistic properties (e.g. lexical, (morpho-)syntactic, semantic), such as those proposed by Beth Levin (1993). However, full exploitation of such classes in real-world tasks has been limited because no comprehensive or domain-specific lexical classification is available. This thesis investigates how Levin-style lexical semantic classes could be learned automatically from corpus data. Automatic acquisition is cost-effective when it involves either no or minimal supervision and it can be applied to any domain of interest where adequate corpus data is available. We improve on earlier work on automatic verb clustering. We introduce new features and new clustering methods to improve the accuracy and coverage. We evaluate our methods and features on well-established cross-domain datasets in English, on a specific domain of English (the biomedical) and on another language (French), reporting promising results. Finally, our task-based evaluation demonstrates that the automatically acquired lexical classes enable new approaches to some NLP tasks (e.g. metaphor identification) and help to improve the accuracy of existing ones (e.g. argumentative zoning).
|
3 |
Multilingual Distributional Lexical SimilarityBaker, Kirk 29 September 2008 (has links)
No description available.
|
4 |
Die valensie van bewegingswerkwoorde in AfrikaansVan der Merwe, Amanda-Marie 06 1900 (has links)
Text in Afrikaans / Twee sentrale probleme word hier ondersoek, naamlik die
daarstelling van 'n teoreties adekwate model om leksikale
valensie te verreken en die problematiek rondom die
beskrywing van Afrikaanse bewegingswerkwoorde.
Die vertrekpunt is dat 'n teorie van UG lesikale valensie
slegs ten dele verreken. So 'n teorie is gemoeid met die vorm
van 'n konstruksie wat op LF-vlak leesbaar is. Gevolglik
beskryf 'n teorie van UG valensie net in universele
sintakties-kategoriale terme. In hierdie studie word die
semantiese en pragmatiese komponente van 'n valensieteorie
derhalwe uitgebou sodat daar 'n duideliker begrip verkry kan
word van die wisselwerking tussen alle veranderlikes van
valensie op 'n taalspesifieke LF-vlak.
'n Valensieteorie word op eklektiese wyse saamgestel uit
verskeie ander teoriee. Die apparaat om die sintaktiese
komponent van 'n valensieteorie te beskryf, word aan TGGmodelle
ontleen. Die semantiese komponent van die teorie word
aangevul vanuit die Konseptuele Semantiek en die pragmatiese
komponent daarvan vanui t kogni ti ewe grammatikamodelle. Die
interaksie tussen hierdie komponente word verreken deur
teoriee van leksikalisasie, korrespondensiereels en die
passing tussen konstruksies en leksikale items.
Die ontleding van Afrikaanse bewegingswerkwoorde spesif iseer
die omvattender anali tiese apparaat wat nodig is vir die
verrekening van valensie verby die punt waarvoor 'n teorie
van UG voorsiening maak. Dit blyk uit hierdie analise dat die
valensie van die kategorie in 'n aantal valensieraambeskrywings
vir die aparte subklasse saamgevat kan word. 'n
Duidelike beeld van die universele en taalspesifiekidiosinkratiese
aspekte van hierdie kategorie kan in terme
van hierdie valensierame aangetoon word.
'n Verdere hipotese (die onakkusatiwiteithipotese) is dat die
sintaktiese verspreiding van werkwoorde ui t hulle semantiek
voorspel kan word. Hierdie hipotese word getoets aan die
empiriese data van Afrikaanse bewegingswerkwoorde. Hierdie
bewegingswerkwoorde word in groepe geklassifiseer na gelang
van ooreenstemmende semanties-pragmatiese kenmerke. Daar word
getoon dat bewegingswerkwoorde met dieself de semantiespragmatiese
kenmerke dieselfde valensierame het. Valensierame
is daarom op sistematiese wyse uit semanties-pragmatiese
kenmerke afleibaar. Sekere fasette van valensierame word
egter ook bepaal deur konvensies soos profilering, en deur
bereelde interaksie met 'n basiese konstruksie se argumentstruktuur. / This thesis focuses on the development of a theoretically
adequate model to account for lexical valence and to provide
the detail for such a model by means of an analysis of
Afrikaans verbs of motion.
The thesis is a response to a theory of UG which inadequately
accounts for lexical valence. A theory of UG is concerned
with the universal syntactic form of a construction that is
legible on LF level. It therefore describes valence in
universal syntactic-categorial terms only. This thesis
develops the semantic and pragmatic components of a theory of
valence in order that a clearer understanding may be gained
of the interaction between all the variables of valence on a
language specific LF level.
A theory of valence is construed eclectically from several
other theories. TGG models provide the methods used to
describe the syntactic component of a theory of valence.
Conceptual Semantics contributes to the development of the
semantic component, and models of cognitive grammar to the
pragmatic component. The interaction between these
components is explained by theories of lexicalisation, rules
of correspondence and the fusion of constructions and lexical
items.
The analysis of Afrikaans verbs of motion provides the
comprehensive set of analytical devices required to account
for valence beyond the point of LF. From this analysis it
emerges that the valence of this category can be summarized
in a finite number of frames of valence for distinct classes
within the category. 'n Clear understanding of the universal
and language specific aspects of this category is achieved by
means of these frames of valence.
A further hypothesis (the unaccusativity hypothesis) posits
that the syntactic distribution of verbs can be predicted
from their semantics. This hypothesis is tested by means of
empirical data of Afrikaans verbs of motion. These verbs are
classified according to similar semantic-pragmatic features.
It is shown that verbs of motion with the same semanticpragmatic
features share frames of valence. These frames thus
are deducible systematically from semantic-pragmatic
features. Certain aspects of frames of valence, however, are
determined by conventions such as profiling and the
regulated interaction between verbs and constructions. / Afrikaans / D.Litt. et Phil. (Afrikaans)
|
5 |
Die valensie van bewegingswerkwoorde in AfrikaansVan der Merwe, Amanda-Marie 06 1900 (has links)
Text in Afrikaans / Twee sentrale probleme word hier ondersoek, naamlik die
daarstelling van 'n teoreties adekwate model om leksikale
valensie te verreken en die problematiek rondom die
beskrywing van Afrikaanse bewegingswerkwoorde.
Die vertrekpunt is dat 'n teorie van UG lesikale valensie
slegs ten dele verreken. So 'n teorie is gemoeid met die vorm
van 'n konstruksie wat op LF-vlak leesbaar is. Gevolglik
beskryf 'n teorie van UG valensie net in universele
sintakties-kategoriale terme. In hierdie studie word die
semantiese en pragmatiese komponente van 'n valensieteorie
derhalwe uitgebou sodat daar 'n duideliker begrip verkry kan
word van die wisselwerking tussen alle veranderlikes van
valensie op 'n taalspesifieke LF-vlak.
'n Valensieteorie word op eklektiese wyse saamgestel uit
verskeie ander teoriee. Die apparaat om die sintaktiese
komponent van 'n valensieteorie te beskryf, word aan TGGmodelle
ontleen. Die semantiese komponent van die teorie word
aangevul vanuit die Konseptuele Semantiek en die pragmatiese
komponent daarvan vanui t kogni ti ewe grammatikamodelle. Die
interaksie tussen hierdie komponente word verreken deur
teoriee van leksikalisasie, korrespondensiereels en die
passing tussen konstruksies en leksikale items.
Die ontleding van Afrikaanse bewegingswerkwoorde spesif iseer
die omvattender anali tiese apparaat wat nodig is vir die
verrekening van valensie verby die punt waarvoor 'n teorie
van UG voorsiening maak. Dit blyk uit hierdie analise dat die
valensie van die kategorie in 'n aantal valensieraambeskrywings
vir die aparte subklasse saamgevat kan word. 'n
Duidelike beeld van die universele en taalspesifiekidiosinkratiese
aspekte van hierdie kategorie kan in terme
van hierdie valensierame aangetoon word.
'n Verdere hipotese (die onakkusatiwiteithipotese) is dat die
sintaktiese verspreiding van werkwoorde ui t hulle semantiek
voorspel kan word. Hierdie hipotese word getoets aan die
empiriese data van Afrikaanse bewegingswerkwoorde. Hierdie
bewegingswerkwoorde word in groepe geklassifiseer na gelang
van ooreenstemmende semanties-pragmatiese kenmerke. Daar word
getoon dat bewegingswerkwoorde met dieself de semantiespragmatiese
kenmerke dieselfde valensierame het. Valensierame
is daarom op sistematiese wyse uit semanties-pragmatiese
kenmerke afleibaar. Sekere fasette van valensierame word
egter ook bepaal deur konvensies soos profilering, en deur
bereelde interaksie met 'n basiese konstruksie se argumentstruktuur. / This thesis focuses on the development of a theoretically
adequate model to account for lexical valence and to provide
the detail for such a model by means of an analysis of
Afrikaans verbs of motion.
The thesis is a response to a theory of UG which inadequately
accounts for lexical valence. A theory of UG is concerned
with the universal syntactic form of a construction that is
legible on LF level. It therefore describes valence in
universal syntactic-categorial terms only. This thesis
develops the semantic and pragmatic components of a theory of
valence in order that a clearer understanding may be gained
of the interaction between all the variables of valence on a
language specific LF level.
A theory of valence is construed eclectically from several
other theories. TGG models provide the methods used to
describe the syntactic component of a theory of valence.
Conceptual Semantics contributes to the development of the
semantic component, and models of cognitive grammar to the
pragmatic component. The interaction between these
components is explained by theories of lexicalisation, rules
of correspondence and the fusion of constructions and lexical
items.
The analysis of Afrikaans verbs of motion provides the
comprehensive set of analytical devices required to account
for valence beyond the point of LF. From this analysis it
emerges that the valence of this category can be summarized
in a finite number of frames of valence for distinct classes
within the category. 'n Clear understanding of the universal
and language specific aspects of this category is achieved by
means of these frames of valence.
A further hypothesis (the unaccusativity hypothesis) posits
that the syntactic distribution of verbs can be predicted
from their semantics. This hypothesis is tested by means of
empirical data of Afrikaans verbs of motion. These verbs are
classified according to similar semantic-pragmatic features.
It is shown that verbs of motion with the same semanticpragmatic
features share frames of valence. These frames thus
are deducible systematically from semantic-pragmatic
features. Certain aspects of frames of valence, however, are
determined by conventions such as profiling and the
regulated interaction between verbs and constructions. / Afrikaans / D.Litt. et Phil. (Afrikaans)
|
Page generated in 0.1854 seconds