The Experts below are selected from a list of 183 Experts worldwide ranked by ideXlab platform

Walter Daelemans - One of the best experts on this subject based on the ideXlab platform.

  • Disambiguation of the neuter pronoun and its effect on pronominal coreference resolution
    Lecture Notes in Computer Science, 2007
    Co-Authors: Veronique Hoste, Iris Hendrickx, Walter Daelemans
    Abstract:

    Coreference resolution, determining the appropriate Discourse Referent for an anaphoric expression, is an essential but difficult task in natural language processing. It has been observed that an important source of errors in machine-learning based approaches to this task, is the wrong disambiguation of the third person singular neuter pronoun as either Referential or non-Referential. In this paper, we investigate whether a machine learning based approach can be successfully applied to the disambiguation of the neuter pronoun in Dutch and show a modest potential effect of this disambiguation on the results of a machine learning based coreference resolution system for Dutch.

  • TSD - Disambiguation of the neuter pronoun and its effect on pronominal coreference resolution
    Text Speech and Dialogue, 1
    Co-Authors: Veronique Hoste, Iris Hendrickx, Walter Daelemans
    Abstract:

    Coreference resolution, determining the appropriate Discourse Referent for an anaphoric expression, is an essential but difficult task in natural language processing. It has been observed that an important source of errors in machine-learning based approaches to this task, is the wrong disambiguation of the third person singular neuter pronoun as either Referential or non-Referential. In this paper, we investigate whether a machine learning based approach can be successfully applied to the disambiguation of the neuter pronoun in Dutch and show a modest potential effect of this disambiguation on the results of a machine learning based coreference resolution system for Dutch.

Paul Dekker - One of the best experts on this subject based on the ideXlab platform.

  • Speaker's Reference, Descriptions and Information Structure
    Journal of Semantics, 1998
    Co-Authors: Paul Dekker
    Abstract:

    The notion of information developed in systems of dynamic semantics is applied in an analysis of the Referential interpretation of definite descriptions, and the specific interpretation of indefinite ones. A Russellian treatment of descriptions is upgraded with the dynamic semantic notion of a Discourse Referent, and this enables a combination of contextually given information with information which is properly semantic. The analysis is sharpened by the addition of a partition of utterances into a ground and a focus part. The two extensions suffice to account for the most important features of situations which involve the Referential use of expressions in both a semantically and pragmatically satisfactory way. A by-product is a rudimentary analysis of negative existential statements involving names

Veronique Hoste - One of the best experts on this subject based on the ideXlab platform.

  • Disambiguation of the neuter pronoun and its effect on pronominal coreference resolution
    Lecture Notes in Computer Science, 2007
    Co-Authors: Veronique Hoste, Iris Hendrickx, Walter Daelemans
    Abstract:

    Coreference resolution, determining the appropriate Discourse Referent for an anaphoric expression, is an essential but difficult task in natural language processing. It has been observed that an important source of errors in machine-learning based approaches to this task, is the wrong disambiguation of the third person singular neuter pronoun as either Referential or non-Referential. In this paper, we investigate whether a machine learning based approach can be successfully applied to the disambiguation of the neuter pronoun in Dutch and show a modest potential effect of this disambiguation on the results of a machine learning based coreference resolution system for Dutch.

  • TSD - Disambiguation of the neuter pronoun and its effect on pronominal coreference resolution
    Text Speech and Dialogue, 1
    Co-Authors: Veronique Hoste, Iris Hendrickx, Walter Daelemans
    Abstract:

    Coreference resolution, determining the appropriate Discourse Referent for an anaphoric expression, is an essential but difficult task in natural language processing. It has been observed that an important source of errors in machine-learning based approaches to this task, is the wrong disambiguation of the third person singular neuter pronoun as either Referential or non-Referential. In this paper, we investigate whether a machine learning based approach can be successfully applied to the disambiguation of the neuter pronoun in Dutch and show a modest potential effect of this disambiguation on the results of a machine learning based coreference resolution system for Dutch.

Katharina Haude - One of the best experts on this subject based on the ideXlab platform.

  • On Discourse-semantic prominence, syntactic prominence, and prominence of expression: the case of Movima
    Journal of Pragmatics, 2019
    Co-Authors: Katharina Haude
    Abstract:

    This study investigates the relationship between three types of linguistic prominence. The first type, Discourse-semantic prominence, encompasses features pertaining to a Discourse Referent at a given moment, like topicality, person, animacy, or agentivity. The second type, syntactic prominence, is reflected by syntactic privileges of an argument. The third type, prominence of linguistic expression, occurs when a linguistic unit stands out through e.g. length and complexity. Cross-linguistically, a Discourse-semantically prominent Referent tends to be syntactically prominent but is encoded by a nonprominent linguistic expression (e.g. an unstressed pronoun). In Movima (isolate, Bolivia), a language with a direct-inverse system, the way the two arguments of a transitive clause are encoded depends on the relative Discourse-semantic prominence of their Referents. Interestingly, the argument representing the less prominent Referent is syntactically privileged, having access to extraction constructions. This unexpected mismatch between Discourse-semantic and syntactic prominence can be explained by the fact that extraction creates a prominent form of linguistic expression, which therefore applies more naturally to an argument with a Discourse-semantically nonprominent Referent. The Movima findings thus show that Discourse-semantic and syntactic prominence do not necessarily correlate, while supporting the assumption that Discourse-semantic prominence universally correlates with low prominence of expression.

  • A topic-marking cleft? Analyzing clause-initial pronouns in Movima
    2018
    Co-Authors: Katharina Haude
    Abstract:

    While basic clauses in Movima are predicate-initial, one construction contains an initial free pronoun. Syntactic tests show that this "pronominal construction", rather than resulting from a word-order change, is syntactically complex: The pronoun serves as the predicate and the verb behaves like a relative clause. Such biclausal structures representing a propositionally equivalent alternative to a basic clause are known as clefts. However, the functions of the pronominal construction contradict those of canonical clefts, which primarily mark argument focus. The free pronoun encodes as a sentence topic a recently introduced Discourse Referent, about which the verb provides an assertion. Therefore, the pronominal construction challenges common definitions of clefts that link a particular structure to a prototypical function.

Agata Renans - One of the best experts on this subject based on the ideXlab platform.

  • Definite descriptions of events: progressive interpretation in Ga (Kwa)
    Linguistics and Philosophy, 2019
    Co-Authors: Agata Renans
    Abstract:

    This paper demonstrates that the progressive interpretation in Ga is an effect of the interaction between the imperfective aspect and a definite description of events. Crucially, the data from Ga point to the consequences of the view that definite descriptions of events encode the familiarity of the Discourse Referent and its uniqueness in bearing the property in question. Namely, they yield direct evidentiality and the necessary ongoingness of the event at the topic time. Thus, the paper identifies previously unattested variation in the semantics of the progressive in a cross-linguistic perspective and shows that not only lexical but also grammatical aspect exhibits striking parallelisms with the nominal domain.