Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 9

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  compositionality
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
We propose a strategy to build the distributional meaning of sentences mainly based on two types of semantic objects: context vectors associated with content words and compositional operations driven by syntactic dependencies. The compositional operations of a syntactic dependency make use of two input vectors to build two new vectors representing the contextualized sense of the two related words. Given a sentence, the iterative application of dependencies results in as many contextualized vectors as content words the sentence contains. At the end of the contextualization process, we do not obtain a single compositional vector representing the semantic denotation of the whole sentence (or of the root word), but one contextualized vector for each constituent word of the sentence. Our method avoids the troublesome high-order tensor representations of approaches relying on category theory, by defining all words as first-order tensors (i.e. standard vectors). Some corpus-based experiments are performed to both evaluate the quality of the contextualized vectors built with our strategy, and to compare them to other approaches on distributional compositional semantics. The experiments show that our dependency-based method performs as (or even better than) the state-of-the-art.
EN
Particle verbs represent a type of multi-word expression composed of a base verb and a particle. The meaning of the particle verb is often, but not always, derived from the meaning of the base verb, sometimes in quite complex ways. In this work, we computationally assess the levels of German particle verb compositionality by applying distributional semantic models. Furthermore, we investigate properties of German particle verbs at the syntax-semantics interface that influence their degrees of compositionality: (i) regularity in semantic particle verb derivation and (ii) transfer of syntactic subcategorization from base verbs to particle verbs. Our distributional models show that both superficial window co-occurrence models as well as theoretically well-founded syntactic models are sensitive to subcategorization frame transfer and can be used to predict degrees of particle verb compositionality, with window models performing better even though they are conceptually and computationally simpler.
3
Content available remote Interacting Workflow Nets for Workflow Process Re-Engineering
EN
In this work we consider modeling of workflow systems with Petri nets. To increase flexibility and give tools for workflow models re-engineering we extend the formalism of workflow nets by considering systems of interacting nets. Then we study soundness – the main correctness property of workflow processes – and show, that for a special class of structured workflow system soundness can be proved in compositional way.
4
Content available remote Associativity of Infinite Synchronized Shuffles and Team Automata
EN
Motivated by different ways to obtain team automata from synchronizing component automata, we consider various definitions of synchronized shuffles of words. A shuffle of two words is an interleaving of their symbol occurrenceswhich preserves the original order of these occurrences within each of the two words. In a synchronized shuffle, however, also two occurrences of one symbol, each from a different word, may be identified as a single occurrence. In case at least one of the words involved is infinite, a (synchronized) shuffle can also be unfair in the sense that an infinite word may prevail fromsome point onwards even when the other word still has occurrences to contribute to the shuffle. We prove that for the synchronized shuffle operations under consideration, every (fair or unfair) synchronized shuffle can be obtained as a limit of synchronized shuffles of the finite prefixes of the words involved. In addition, it is shown that with the exception of one, all synchronized shuffle operations that we consider satisfy a natural notion of associativity, also in case of unfairness. Finally, using these results, some compositionality results for team automata are established.
5
Content available remote Discourse interpretation based on dynamic constraints
EN
Our main objective will be to construct a fully compositional representation of nominal anaphora in discourse. The proposed representation is not dependent on the remote ascription (i.e. done outside the formal representation) of syntactic indexes, which identify anaphoric links. A formal language of variable free logic is introduced. It is based on dynamic semantics paradigm and is a variant of many-sorted type logic. We will also present the scope free treatment of quantification in multiple quantifier sentences. The interpretation of multiple quantifiers is defined by means of a construction of the polyadic Generalised Quantifier (GQ). The polyadic GQ is a constraint that should be satisfied by the denotation of a ?clausal? predicate.
6
Content available remote Relating Communicating Processes with Different Interfaces
EN
We present here an implementation relation intended to formalise the notion that a system built of communicating processes is an acceptable implementation of another base, or target, system in the event that the two systems have different interfaces. Such a treatment has clear applicability in the software development process, where (the interface of) an implementation component may be expressed at a different level of abstraction to (the interface of) the relevant specification component. Technically, processes are formalised using Hoare's CSP language, with its standard failures-divergences model. The implementation relation is formulated in terms of failures and divergences of the implementation and target processes. Interface difference is modelled by endowing the implementation relation with parameters called extraction patterns. These are intended to interpret implementation behaviour as target behaviour, and suitably constrain the former in connection to well-formedness and deadlock properties. We extend the results of our previous work and replace implementation relations previously presented by a single, improved scheme. We also remove all the restrictions previously placed upon target processes. Two basic kinds of results are obtained: realisability and compositionality. The latter means that a target composed of several connected systems may be implemented by connecting their respective implementations. The former means that, if target and implementation in fact have the same interface, then the implementation relation they should satisfy collapses into standard implementation pre-order. We also show how to represent processes and extraction patterns in a manner amenable to computer implementation, and detail a graph-theoretic restatement of the conditions defining the implementation relation, from which algorithms for their automatic verification are easily derived.
EN
This paper presents an extension of an algebra of high-level Petri nets with operations for suspension and abortion. These operations are sound with respect to the semantics of preemption, and can be applied to the modelling of the semantics of high-level parallel programming languages with preemption-related features. As an illustration, the paper gives an application to the modelling of a multi-tasking system in a parallel programming language, which is provided with a concurrent semantics based on Petri nets and for which implemented tools can be used.
8
Content available remote Behaviour Abstraction for Communicating Sequential Processes
EN
We investigate the notion that a system is an acceptable implementation of another base or target system, in the case that the two systems (or processes) have different interfaces. For instance, the base processes can be thought of as specifications, or ideal processes operating in an error-free environment, while the implementations model their actual realisations, operating in an error-prone environment and possibly employing a variety of fault-tolerant techniques. Using the CSP model, we relate implementations and base systems in terms of their observable behaviours. We obtain two fundamental results, viz. realisability and compositionality. The former ensures that implementations may be put to good use, while the latter guarantees that a target composed of several connected subsystems may be implemented by connecting their respective implementations.
9
Content available remote Integrating Distributed Algorithms into distributed systems
EN
A distributed algorithm is often used as a part of a larger distributed system. Usually, the properties of an algorithm are proven for the algorithm in isolation. Then, it is not obvious how the algorithm behaves when integrated into a larger system. In this paper, we present a simple technique which allows to derive properties of an algorithm which is integrated into a distributed system from the properties of the algorithm in isolation. The technique exploits the fact that some actions of a distributed algorithm do not belong to the algorithm but are triggered by the environment. If these actions are distinguished and are adequately considered in the verification of the algorithm, basically all properties are still valid for the algorithm as a part of a larger distributed system. This result will be formalized in the setting of the Distributed Algorithms' Working Notation (DAWN). Based on this result, we will give a proof rule which allows to prove liveness properties of a system from liveness properties of the involved distributed algorithm.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.