The Experts below are selected from a list of 69561 Experts worldwide ranked by ideXlab platform

Fabrizio Santini - One of the best experts on this subject based on the ideXlab platform.

  • ICANN - Recursive Neural Networks Applied to Discourse Representation Theory
    Artificial Neural Networks — ICANN 2002, 2002
    Co-Authors: Antonella Bua, Marco Gori, Fabrizio Santini
    Abstract:

    Connectionist semantic modeling in natural language processing (a typical symbolic domain) is still a challenging problem. This paper introduces a novel technique, combining Discourse Representation Theory (DRT) with Recursive Neural Networks (RNN) in order to yield a neural model capable to discover properties and relationships among constituents of a knowledge-base expressed by natural language sentences. DRT transforms sequences of sentences into directed ordered acyclic graphs, while RNNs are trained to deal with such structured data. The acquired information allows the network to reply on questions, the answers of which are not directly expressed into the knowledge-base. A simple experimental demonstration, drawn from the context of a fairy tales is presented. Finally, on-going research direction are pointed-out.

  • Recursive Neural Networks applied to Discourse Representation Theory
    Lecture Notes in Computer Science, 2002
    Co-Authors: Antonella Bua, Marco Gori, Fabrizio Santini
    Abstract:

    Connectionist semantic modeling in natural language processing (a typical symbolic domain) is still a challenging problem. This paper introduces a novel technique, combining Discourse Representation Theory (DRT) with Recursive Neural Networks (RNN) in order to yield a neural model capable to discover properties and relationships among constituents of a knowledge-base expressed by natural language sentences. DRT transforms sequences of sentences into directed ordered acyclic graphs, while RNNs are trained to deal with such structured data. The acquired information allows the network to reply on questions, the answers of which are not directly expressed into the knowledge-base. A simple experimental demonstration, drawn from the context of a fairy tales is presented. Finally, on-going research direction are pointed-out.

Johan Bos - One of the best experts on this subject based on the ideXlab platform.

  • DRS at MRP 2020: Dressing up Discourse Representation Structures as Graphs.
    arXiv: Computation and Language, 2020
    Co-Authors: Lasha Abzianidze, Johan Bos, Stephan Oepen
    Abstract:

    Discourse Representation Theory (DRT) is a formal account for representing the meaning of natural language Discourse. Meaning in DRT is modeled via a Discourse Representation Structure (DRS), a meaning Representation with a model-theoretic interpretation, which is usually depicted as nested boxes. In contrast, a directed labeled graph is a common data structure used to encode semantics of natural language texts. The paper describes the procedure of dressing up DRSs as directed labeled graphs to include DRT as a new framework in the 2020 shared task on Cross-Framework and Cross-Lingual Meaning Representation Parsing. Since one of the goals of the shared task is to encourage unified models for several semantic graph frameworks, the conversion procedure was biased towards making the DRT graph framework somewhat similar to other graph-based meaning Representation frameworks.

  • CoNLL Shared Task - DRS at MRP 2020: Dressing up Discourse Representation Structures as Graphs
    Proceedings of the CoNLL 2020 Shared Task: Cross-Framework Meaning Representation Parsing, 2020
    Co-Authors: Lasha Abzianidze, Johan Bos, Stephan Oepen
    Abstract:

    Discourse Representation Theory (DRT) is a formal account for representing the meaning of natural language Discourse. Meaning in DRT is modeled via a Discourse Representation Structure (DRS), a meaning Representation with a model-theoretic interpretation, which is usually depicted as nested boxes. In contrast, a directed labeled graph is a common data structure used to encode semantics of natural language texts. The paper describes the procedure of dressing up DRSs as directed labeled graphs to include DRT as a new framework in the 2020 shared task on Cross-Framework and Cross-Lingual Meaning Representation Parsing. Since one of the goals of the shared task is to encourage unified models for several semantic graph frameworks, the conversion procedure was biased towards making the DRT graph framework somewhat similar to other graph-based meaning Representation frameworks.

  • The First Shared Task on Discourse Representation Structure Parsing
    Proceedings of the IWCS Shared Task on Semantic Parsing, 2019
    Co-Authors: Lasha Abzianidze, Rik Van Noord, Hessel Haagsma, Johan Bos
    Abstract:

    The paper presents the IWCS 2019 shared task on semantic parsing where the goal is to produce Discourse Representation Structures (DRSs) for English sentences. DRSs originate from Discourse Representation Theory and represent scoped meaning Representations that capture the semantics of negation, modals, quantification, and presupposition triggers. Additionally, concepts and event-participants in DRSs are described with WordNet synsets and the thematic roles from VerbNet. To measure similarity between two DRSs, they are represented in a clausal form, i.e. as a set of tuples. Participant systems were expected to produce DRSs in this clausal form. Taking into account the rich lexical information, explicit scope marking, a high number of shared variables among clauses, and highly-constrained format of valid DRSs, all these makes the DRS parsing a challenging NLP task. The results of the shared task displayed improvements over the existing state-of-the-art parser.

  • Exploring Neural Methods for Parsing Discourse Representation Structures
    arXiv: Computation and Language, 2018
    Co-Authors: Rik Van Noord, Lasha Abzianidze, Antonio Toral, Johan Bos
    Abstract:

    Neural methods have had several recent successes in semantic parsing, though they have yet to face the challenge of producing meaning Representations based on formal semantics. We present a sequence-to-sequence neural semantic parser that is able to produce Discourse Representation Structures (DRSs) for English sentences with high accuracy, outperforming traditional DRS parsers. To facilitate the learning of the output, we represent DRSs as a sequence of flat clauses and introduce a method to verify that produced DRSs are well-formed and interpretable. We compare models using characters and words as input and see (somewhat surprisingly) that the former performs better than the latter. We show that eliminating variable names from the output using De Bruijn-indices increases parser performance. Adding silver training data boosts performance even further.

  • ugroningen negation detection with Discourse Representation structures
    Joint Conference on Lexical and Computational Semantics, 2012
    Co-Authors: Valerio Basile, Johan Bos, Kilian Evang, Noortje Venhuizen
    Abstract:

    We use the NLP toolchain that is used to construct the Groningen Meaning Bank to address the task of detecting negation cue and scope, as defined in the shared task "Resolving the Scope and Focus of Negation". This toolchain applies the C&C tools for parsing, using the formalism of Combinatory Categorial Grammar, and applies Boxer to produce semantic Representations in the form of Discourse Representation Structures (DRSs). For negation cue detection, the DRSs are converted to flat, non-recursive structures, called Discourse Representation Graphs (DRGs). DRGs simplify cue detection by means of edge labels representing relations. Scope detection is done by gathering the tokens that occur within the scope of a negated DRS. The result is a system that is fairly reliable for cue detection and scope detection. Furthermore, it provides a fairly robust algorithm for detecting the negated event or property within the scope.

Antonella Bua - One of the best experts on this subject based on the ideXlab platform.

  • ICANN - Recursive Neural Networks Applied to Discourse Representation Theory
    Artificial Neural Networks — ICANN 2002, 2002
    Co-Authors: Antonella Bua, Marco Gori, Fabrizio Santini
    Abstract:

    Connectionist semantic modeling in natural language processing (a typical symbolic domain) is still a challenging problem. This paper introduces a novel technique, combining Discourse Representation Theory (DRT) with Recursive Neural Networks (RNN) in order to yield a neural model capable to discover properties and relationships among constituents of a knowledge-base expressed by natural language sentences. DRT transforms sequences of sentences into directed ordered acyclic graphs, while RNNs are trained to deal with such structured data. The acquired information allows the network to reply on questions, the answers of which are not directly expressed into the knowledge-base. A simple experimental demonstration, drawn from the context of a fairy tales is presented. Finally, on-going research direction are pointed-out.

  • Recursive Neural Networks applied to Discourse Representation Theory
    Lecture Notes in Computer Science, 2002
    Co-Authors: Antonella Bua, Marco Gori, Fabrizio Santini
    Abstract:

    Connectionist semantic modeling in natural language processing (a typical symbolic domain) is still a challenging problem. This paper introduces a novel technique, combining Discourse Representation Theory (DRT) with Recursive Neural Networks (RNN) in order to yield a neural model capable to discover properties and relationships among constituents of a knowledge-base expressed by natural language sentences. DRT transforms sequences of sentences into directed ordered acyclic graphs, while RNNs are trained to deal with such structured data. The acquired information allows the network to reply on questions, the answers of which are not directly expressed into the knowledge-base. A simple experimental demonstration, drawn from the context of a fairy tales is presented. Finally, on-going research direction are pointed-out.

Reinhard Muskens - One of the best experts on this subject based on the ideXlab platform.

  • Construction by Description in Discourse Representation
    2003
    Co-Authors: N. Van Leusen, Reinhard Muskens
    Abstract:

    This paper uses classical logic for a simultaneous description of the syntax and semantics of a fragment of English and it is argued that such an approach to natural language allows procedural aspects of linguistic theory to get a purely declarative formulation. In particular, it will be shown how certain construction rules in Discourse Representation Theory, such as the rule that indenites create new Discourse referents and denites pick up an existing referent, can be formulated declaratively if logic is used as a metalanguage for English. In this case the declarative aspects of a rule are highlighted when we focus on the model theory of the description language while a procedural perspective is obtained when its proof theory is concentrated on. Themes of interest are Discourse Representation Theory, resolution of anaphora, resolution of presuppositions, and underspecication.

  • Chapter 10 – Dynamics
    Handbook of Logic and Language, 1997
    Co-Authors: Reinhard Muskens
    Abstract:

    Publisher Summary This chapter provides an overview of some important dynamic theories in linguistics and artificial intelligence and shows the way in which these fit into the general perspective on communication. It contains some more general logical considerations on dynamic phenomena, discussing various ways to model their logic and the way the logic that emerges is related to its classical static predecessors. Natural language involves different dynamic mechanisms. Discourse Representation theory and dynamic predicate logic highlight changing anaphoric bindings, whereas Veltman's Update semantics focuses on information flow and epistemic statements about its stages. The basic ideas of Heim's file change semantics (FCS) and Kamp's Discourse Representation theory (DRT) are very much the same. While in Heim's theory the readers or hearers of a text represent the information that they have obtained by means of a file, DRT lets them keep track of that information with the help of a Discourse Representation structure.

  • Combining Montague semantics and Discourse Representation
    Linguistics and Philosophy, 1996
    Co-Authors: Reinhard Muskens
    Abstract:

    This paper embeds the core part of Discourse Representation Theory in the classical theory of types plus a few simple axioms that allow the theory to express key facts about variables and assignments on the object level of the logic. It is shown how the embedding can be used to combine core analyses of natural language phenomena in Discourse Representation Theory with analyses that can be obtained in Montague Semantics.

  • Negation and disjunction in Discourse Representation theory
    Journal of Semantics, 1995
    Co-Authors: Emiel Krahmer, Reinhard Muskens
    Abstract:

    La Theorie de la Representation du Discours classique (TRD) predit qu'un SN indefini ne peut preceder un element anaphorique si le SN est -et l'element anaphorique n'est pas- dans le champ de la negation. La theorie predit egalement qu'aucun lien anaphorique n'est possible entre les deux parties de la disjonction. Il existe toutefois des contre-exemples a ces predictions. Les As. examinent ces contre-exemples et proposent une extension a la TRD. Ils esquissent une adaptation de la construction algorithmique de la TRD standard, qui transforme les textes en Structures de Representation du Discours

  • COLING - Categorial Grammar and Discourse Representation Theory
    Proceedings of the 15th conference on Computational linguistics -, 1994
    Co-Authors: Reinhard Muskens
    Abstract:

    In this paper it is shown how simple texts that can be parsed in a Lambek Categorial Grammar can also automatically be provided with a semantics in the form of a Discourse Representation Structure in the sense of Kamp [1981]. The assignment of meanings to texts uses the Curry-Howard-Van Benthem correspondence.

Noortje Venhuizen - One of the best experts on this subject based on the ideXlab platform.

  • PDRT-SANDBOX: An implementation of Projective Discourse Representation Theory
    2014
    Co-Authors: Noortje Venhuizen, Harm Brouwer
    Abstract:

    We introduce PDRT-SANDBOX, a Haskell library that implements Projective Dis- course Representation Theory (PDRT) (Venhuizen et al., 2013), an extension of Discourse Representation Theory (DRT) (Kamp, 1981; Kamp and Reyle, 1993). The implementation includes a translation from PDRT to DRT and first-order logic, composition via different types of merge, and unresolved structures based on Mon- tague Semantics (Muskens, 1996), defined as Haskell functions.

  • PDRT-SANDBOX : A Haskell NLP library implementing DRT and PDRT
    2014
    Co-Authors: Harm Brouwer, Noortje Venhuizen
    Abstract:

    PDRT-SANDBOX is a Haskell NLP library that implements Discourse Representation Theory (DRT), and its extension Projective Discourse Representation Theory (PDRT).

  • ugroningen negation detection with Discourse Representation structures
    Joint Conference on Lexical and Computational Semantics, 2012
    Co-Authors: Valerio Basile, Johan Bos, Kilian Evang, Noortje Venhuizen
    Abstract:

    We use the NLP toolchain that is used to construct the Groningen Meaning Bank to address the task of detecting negation cue and scope, as defined in the shared task "Resolving the Scope and Focus of Negation". This toolchain applies the C&C tools for parsing, using the formalism of Combinatory Categorial Grammar, and applies Boxer to produce semantic Representations in the form of Discourse Representation Structures (DRSs). For negation cue detection, the DRSs are converted to flat, non-recursive structures, called Discourse Representation Graphs (DRGs). DRGs simplify cue detection by means of edge labels representing relations. Scope detection is done by gathering the tokens that occur within the scope of a negated DRS. The result is a system that is fairly reliable for cue detection and scope detection. Furthermore, it provides a fairly robust algorithm for detecting the negated event or property within the scope.

  • *SEM@NAACL-HLT - UGroningen: Negation detection with Discourse Representation Structures
    2012
    Co-Authors: Valerio Basile, Johan Bos, Kilian Evang, Noortje Venhuizen
    Abstract:

    We use the NLP toolchain that is used to construct the Groningen Meaning Bank to address the task of detecting negation cue and scope, as defined in the shared task "Resolving the Scope and Focus of Negation". This toolchain applies the C&C tools for parsing, using the formalism of Combinatory Categorial Grammar, and applies Boxer to produce semantic Representations in the form of Discourse Representation Structures (DRSs). For negation cue detection, the DRSs are converted to flat, non-recursive structures, called Discourse Representation Graphs (DRGs). DRGs simplify cue detection by means of edge labels representing relations. Scope detection is done by gathering the tokens that occur within the scope of a negated DRS. The result is a system that is fairly reliable for cue detection and scope detection. Furthermore, it provides a fairly robust algorithm for detecting the negated event or property within the scope.