Informacja

Drogi użytkowniku, aplikacja do prawidłowego działania wymaga obsługi JavaScript. Proszę włącz obsługę JavaScript w Twojej przeglądarce.

Wyszukujesz frazę "syntactic dependency" wg kryterium: Temat


Wyświetlanie 1-2 z 2
Tytuł:
On different approaches to syntactic analysis into bi-lexical dependencies : An empirical comparison of direct, PCFG-based, and HPSG-based parsers
Autorzy:
Ivanova, A.
Oepen, S.
Dridan, R.
Flickinger, D.
Øvrelid, L.
Lapponi, E.
Tematy:
syntactic dependency parsing
domain variation
Pokaż więcej
Wydawca:
Polska Akademia Nauk. Instytut Podstaw Informatyki PAN
Powiązania:
https://bibliotekanauki.pl/articles/103851.pdf  Link otwiera się w nowym oknie
Opis:
We compare three different approaches to parsing into syntactic, bilexical dependencies for English: a ‘direct’ data-driven dependenci parser, a statistical phrase structure parser, and a hybrid, ‘deep’ grammar-driven parser. The analyses from the latter two are postconverted to bi-lexical dependencies. Through this ‘reduction’ of All three approaches to syntactic dependency parsers, we determine empirically what performance can be obtained for a common set of dependenci types for English; in- and out-of-domain experimentation ranges over diverse text types. In doing so, we observe what trade-offs apply along three dimensions: accuracy, efficiency, and resilience to domain variation. Our results suggest that the hand-built grammar in one of our parsers helps in both accuracy and cross-domain parsing performance. When evaluated extrinsically in two downstream tasks – negation resolution and semantic dependency parsing – these accuracy gains do sometimes but not always translate into improved end-to-end performance.
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
The expected sum of edge lengthsin planar linearizations of trees
Autorzy:
Alemany-Puig, Lluís
Ferrer-i-Cancho, Ramon
Tematy:
dependency grammar
projectivity
planarity
syntactic dependency distance minimization
Pokaż więcej
Wydawca:
Polska Akademia Nauk. Instytut Podstaw Informatyki PAN
Powiązania:
https://bibliotekanauki.pl/articles/59112625.pdf  Link otwiera się w nowym oknie
Opis:
Dependency trees have proven to be a very successful model to rep-resent the syntactic structure of sentences of human languages. Inthese structures, vertices are words and edges connect syntactically-dependent words. The tendency of these dependencies to be short hasbeen demonstrated using random baselines for the sum of the lengthsof the edges or their variants. A ubiquitous baseline is the expectedsum in projective orderings (wherein edges do not cross and the rootword of the sentence is not covered by any edge), that can be com-puted in timeO(n). Here we focus on a weaker formal constraint,namely planarity. In the theoretical domain, we present a characteri-zation of planarity that, given a sentence, yields either the number ofplanar permutations or an efficient algorithm to generate uniformlyrandom planar permutations of the words. We also show the relation-ship between the expected sum in planar arrangements and the ex-pected sum in projective arrangements. In the domain of applications,we derive aO(n)-time algorithm to calculate the expected value ofthe sum of edge lengths. We also apply this research to a parallel cor-pus and find that the gap between actual dependency distance and therandombaselinereducesasthestrengthoftheformalconstraintonde-pendency structures increases, suggesting that formal constraints ab-sorbpartofthedependencydistanceminimizationeffect.Ourresearchpaves the way for replicating past research on dependency distanceminimization using random planar linearizations as random baseline.
Dostawca treści:
Biblioteka Nauki
Artykuł
    Wyświetlanie 1-2 z 2

    Ta witryna wykorzystuje pliki cookies do przechowywania informacji na Twoim komputerze. Pliki cookies stosujemy w celu świadczenia usług na najwyższym poziomie, w tym w sposób dostosowany do indywidualnych potrzeb. Korzystanie z witryny bez zmiany ustawień dotyczących cookies oznacza, że będą one zamieszczane w Twoim komputerze. W każdym momencie możesz dokonać zmiany ustawień dotyczących cookies