Lexicon-Grammar continuum

In this paper we propose an approach for identifying syntactic behaviours related to lexical items and linking them to the meanings. This approach is based on the analysis of the textual content presented in LMF normalized dictionaries by means of Definition and Context classes. The main particularity of these contents is their large availability and their semantically control due to their attachment to the meanings, which promotes the effective links between the syntactic behaviours and the meaning. In order to test the performance of the proposed approach, we tested it on an available Arabic LMF normalized dictionary. The experiment treats 9,800 verbs and allows us to evaluate the identified syntactic behaviours as well as their links to the meanings.

This article investigates the use of negative interrogatives in English and provides new support as to why they can be regarded as argumentative structures (Heritage, 2002). Questions are usually described pragmatically as enabling the speaker to seek information. However, when they are negatively formulated, they are analysed in the literature as allowing the speaker to express their point of view: “negative interrogatives are treated as accomplishing assertions of opinion rather than questioning” (Heritage, 2002). This paper builds on Heritage’s claim by considering the whole discursive project of the speaker.  The rhetorical trait of these structures will necessarily be dealt with. The corpus is comprised of negative interrogatives from the Santa Barbara Corpus of Spoken American English [1] . By analysing the responses that follow questions (Leon, 1997), we show that adding negation to the usual (i.e. positive) interrogative form turns the classical information-seeking question ...

Les tests que nous présentons sont presque tous empruntés à Anscombre (1990), Cadiot (1992) et Gross (1996). 1.3.1. Test de prédicativité (verbe être) (Anscombre 1990, Cadiot 1992) à/POUR ou SA Une brosse à dents-*Cette brosse est à dents Un sac à dos-*Ce sac est à dos Un moulin à vent-*Ce moulin est à vent à/AVEC Un fauteuil à roulettes-??Ce fauteuil est à roulettes Un verre à pied-??Ce verre est à pied Un bateau à voiles-??Ce bateau est à voiles 1.3.2. Test de la paraphrase à l'aide des composants (Anscombre 1990) à/POUR ou SA Une brosse à dents est une (petite) brosse qui sert à nettoyer les dents Un sac à dos est un sac qui est porté sur le dos Un moulin à vent est un moulin qui fonctionne avec le vent à/AVEC Un fauteuil à roulettes est un fauteuil qui a des roulettes Un verre à pied est un verre qui a un pied Un bateau à voiles est un bateau (i) qui a des voiles (ii) qui marche avec des voiles

À partir d'un dictionnaire électronique de plus de 15 000 adjectifs en langue espagnole, que nous avons élaboré dans le cadre du laboratoire fLexSem (Phonétique, Lexicologie et Sémantique, UAB), nous décrivons dans cet article tous les cas où un adjectif unilexical espagnol doit être mis en correspondance avec un équivalent de traduction polylexical en français. L'objectif de cette description est double : d'une part, il s'agit de préciser la notion d'adjectif polylexical en tant que catégorie nécessaire pour établir des équivalences de traduction dans le dictionnaire bilingue. D'autre part, il s'agit aussi d'offrir une typologie formelle complète de ce type d'équivalents de traduction.

In this paper we propose an approach for identifying syntactic behaviours related to lexical items and linking them to the meanings. This approach is based on the analysis of the textual content presented in LMF normalized dictionaries by means of Definition and Context classes. The main particularity of these contents is their large availability and their semantically control due to their attachment to the meanings, which promotes the effective links between the syntactic behaviours and the meaning. In order to test the performance of the proposed approach, we tested it on an available Arabic LMF normalized dictionary. The experiment treats 9,800 verbs and allows us to evaluate the identified syntactic behaviours as well as their links to the meanings.

Adopting the concept of "Local Grammars" (M. Gross), which were successfully applied in practice by (Geierhos, 2010) to biographical information extraction in English our project aims to detect, encode, and finally visualize relations between persons. Our corpus consists of the digitised biographical lexicon "Neue Deutsche Biographie (NDB)", roughly 21.000 biographies in 25 volumes in print since 1953. We developed local grammars and suitable dictionaries to describe interpersonal relations and applied them to the corpus with Unitex 3.1. The local grammars were designed to integrate existing TEI-XML structures in the corpus. Using the ability of local grammars in Unitex to act as transducers we were able to produce XML-tags and encode semantic information. Based on grammars for personal names and places we described interpersonal relations like to study, predecessors and successors as well as friends and circles. Afterwards we identified persons (as given in the authority file or index). Finally we displayed relations on our website in an interactive and dynamic way. Utilizing the Javascript library D3.js we represented named relations between identified individuals as ego centred network graphs.

Pattern-matching finite-state techniques applied to lexical analysis of multiword expressions have received a significant boost from the possibility of intersecting lexical information encoded in lexicon-grammar matrices and reference graphs. However, these methods show ...

We present a method for enlarge a lexicon (with frequencies information), that is useful for parsing and others NLP applications. We show an example enlarging the verbal LGLex lexicon of French [8], using several corpora extracted from the evaluation campaign for French parsers Passage [5]. To do that, we use the results of the frmg parser [7] with IRASubcat [1], a tool that automatically acquires subcategorization frames from corpus in any language and that also allows to complete an existing lexicon. We obtain the frequencies of occurrence for each input and each subcategorization frame for 14,068 distinct lemmas.

This paper describes an experiment on the identification of frozen sentences (or verbal idioms) from European Portuguese on large corpus of journalistic text. It aims at identifying the main difficulties (or shortcomings) resulting from the intersection of linguistic information ...

L'objectif du présent travail est de mener une étude comparée sur l'adjectif et l'adjectivité dans les langues française et arabe, d'observer les tendances générales, et de mettre en lumière d'éventuelles divergences susceptibles de caractériser du point de vue macroscopique le fonctionnement de l'adjectif en tant que catégorie et en tant que constituant de la phrase dans une perspective contrastive. La dimension comparative de notre étude proviendra de la confrontation des deux systèmes linguistiques. C'est ce double examen intralinguistique et interlinguistique qui nous permettrait effectivement de formuler quelques conjectures théoriques, d'entrevoir les différences potentielles entre les deux langues, mais surtout de mettre en évidence l'intérêt d'élargir la perspective d'analyse dans la description des langues.

Distribution électronique Cairn.info pour Publications linguistiques. © Publications linguistiques. Tous droits réservés pour tous pays. La reproduction ou représentation de cet article, notamment par photocopie, n'est autorisée que dans les limites des conditions générales d'utilisation du site ou, le cas échéant, des conditions générales de la licence souscrite par votre établissement. Toute autre reproduction ou représentation, en tout ou partie, sous quelque forme et de quelque manière que ce soit, est interdite sauf accord préalable et écrit de l'éditeur, en dehors des cas prévus par la législation en vigueur en France. Il est précisé que son stockage dans une base de données est également interdit.

This study describes a methodology we adopted for constructing Multilingual Sentiment-Annotated Corpora (named MUSE), that consist of two types of annotated corpora: Sentence-based Sentiment-Annotated Corpora (MUSE-SESAC) and Token-based Sentiment-Annotated Corpora (MUSE-TOSAC).

La sous-catégorisation d'arguments introduits par la préposition pour a été sous-étudiée par le passé, comme en témoigne l'incomplétude des ressources lexico-syntaxiques existantes sur ce point. Dans cet article, nous présentons rapidement les différents types de sous-catégorisation en pour, qui contrastent avec les emplois de pour comme connecteur de discours. Nous décrivons l'intégration des arguments en pour au lexique syntaxique Lefff , enrichissant ainsi les informations de sous-catégorisation de nombreuses entrées verbales, nominales, adjectivales et adverbiales.

We have brought together references of published works by specialists of various institutions, all connected by one same project: the coordinated construction of electronic dictionaries and grammars of different languages. This project, called RELEX, was born at the Laboratoire d'Automatique Documentaire et Linguistique, where linguists, each working in their domain with common theoretical options, have started accumulating com puterized data that constitute coherent lexicon-grammars. This bibliography may have errors and gaps. It is intended to be updated periodically. We would be grateful to our readers for helping us improve it. Abeillé, Anne. 1987. Grammaire des noms de métier.

El següent treball té com a objectiu examinar els contextos més comuns en què es poden trobar clàusules subordinades de that i identificar els factors que limiten l'omissió del that. Aquesta dissertació se centra tant en les restriccions sintàctiques com lèxiques que es poden analitzar en aquests contextos: Verb + Clàusula that, Substantiu + Clàusula that i Adjectiu + Clàusula that. A més, aquest estudi pretén revisar els factors pragmàtics que restringeixen o milloren l'omissió del complementador. Finalment, s'ha conclòs que el fenomen de la supressió no és opcional, ja que està motivat per factors gramaticals i pragmàtics.The present work aims to examine the most common contexts in which that- subordinate clauses can be found and identify the factors that constraint the omission of the complementiser. This dissertation focuses on both the syntactic and lexical restrictions that can be analysed in these contexts: Verb + That Clause, Noun + That Clause and Adjective + Th...

Cette étude est le premier volet d'une étude plus vaste consacrée à l'étude des préfixes dits négatifs. Nous nous bornerons ici à évoquer le cas de in-dans le cas de la formation d'adjectifs. Cette dénomination de « préfixe négatif » vient de ce que dans un nombre appréciable de cas, il existe une paire de type (adjectif en in-, adjectif) telle que : être adjectif en in-= ne pas être adjectif. Ainsi : (attendu, inattendu), (habité, inhabité), (possible, impossible), (stable, instable), (traduisible, intraduisible), et bien d'autres encore. Position classique fort clairement exprimée dans Togeby ; 1985 : « ... Les deux préfixes négatifs in-et anti-s'ajoutent uniquement à des noms (substantifs ou adjectifs)... in-se joint surtout à des adjectifs. Adjectifs dérivés en -able et -ible : incroyable, intraduisible ; participes passés : inattendu, insoupçonné ; adjectifs verbaux en -ant : impuissant, incroyant, insouciant ; autres adjectifs dérivés : impersonnel, infructueux. Il existe aussi des adjectifs simples qui admettent le préfixe in-: indigne, inégal, injuste, impur, etc., mais les adjectifs les plus fréquents, qui forment des couples d'antonymes, ne l'admettent jamais : bon-mauvais, grand-petit, haut-bas, large-étroit, long-court, vraifaux, beau-laid... » (p. 15).Cette position fait immédiatement surgir un certains nombre d'interrogations. On peut par exemple se demander pourquoi les sujets parlants identifient sans hésiter le inde impotent comme ce même préfixe négatif, alors que l'on n'a pas potent. Invoquer l'emprunt (effectif) au latin n'explique rien. Pourquoi ne pas avoir emprunté les deux termes (comme dans le cas de docile/indocile), puisque potens existe en latin, et que par ailleurs, potent apparaît dans certains composés (omnipotent, idempotent, nihilpotent, plénipotentiaire) ? L'insoutenable légèreté morphologique du préfixe négatif in-dans la formatio...

Pattern-matching finite-state techniques applied to lexical analysis of multiword expressions have received a significant boost from the possibility of intersecting lexical information encoded in lexicon-grammar matrices and reference graphs. However, these methods show ...

L'objectif du présent travail est de mener une étude comparée sur l'adjectif et l'adjectivité dans les langues française et arabe, d'observer les tendances générales, et de mettre en lumière d'éventuelles divergences susceptibles de caractériser du point de vue macroscopique le fonctionnement de l'adjectif en tant que catégorie et en tant que constituant de la phrase dans une perspective contrastive. La dimension comparative de notre étude proviendra de la confrontation des deux systèmes linguistiques. C'est ce double examen intralinguistique et interlinguistique qui nous permettrait effectivement de formuler quelques conjectures théoriques, d'entrevoir les différences potentielles entre les deux langues, mais surtout de mettre en évidence l'intérêt d'élargir la perspective d'analyse dans la description des langues.

The development of computational linguistic resources (electronic dictionaries and grammars) for the automatic extraction, identification, and further fine-grained annotation of "complex text segments" , is the core of our work. We use and extend the notion of multi-word units (MWUs) by allowing a large description of linguistic objects: compound nouns, entity names, verbal forms (compound tense and negate forms, introduction of clauses between the auxiliary and the past participle, etc.) and frozen expressions (i.e. idioms). The identification of complex sequences of text segments is done by using dictionary graphs which combines the power and versatility of the local grammars and the expressivity of the electronic dictionaries.

The present research exploits the large amount of linguistic resources developed into the Lexicon-grammar paradigm in the domain of the Opinion Mining. Grounded on the Semantic Predicates theory, the proposed system is able to automatically match the syntactic structures selected by special classes of verbs, indicating positive or negative Sentiment, Opinion or Physical acts, with the semantic frames evoked by the same lexical items. This methods has been tested on a large dataset composed of short texts, such as tweets and news headings.

Les verbes peuvent etre categorises dans des classes homogenes, a la fois du point de vue semantique et du point de vue syntaxique. Chaque classe offre une serie de proprietes communes : schema d'arguments, type semantique (action, etat, evenement), adverbiaux appropries, reconstructions specifiques, proprietes morphologiques. Cette factorisation permet une description formalisee du fonctionnement des verbes et un traitement plus precis de la polysemie.

The purpose of this paper is to approach the problem of disambiguation of one type of polysemic units, that is adjectives in the context of translation. On the basis of the French adjective piquant analyzed as a case of study, following the object oriented method, the author demonstrates that this item could have a few equivalents in other languages. The selection of an appropriate linguistic equivalent depends on the context in which a particular unit appears. Such multiple possible interpretations pose a major problem for the description of the polysemic terms in lexicographical publications, as well as for automatic translation technology. To exactly pinpoint all the available senses and equivalents, the author makes use of the theory of classes of objects.

Polisemia przymiotnikow - wyzwanie dla tlumaczenia automatycznego Polisemia przymiotnikow, jak i innych jednostek jezykowych stanowi wyzwanie dla tlumaczenia, szczegolnie dla tlumaczenia automatycznego. Aby system byl w stanie wykryc odpowiedni ekwiwalent w jezyku docelowym, niezbedny jest opis formalny, biorący pod uwage wszystkie uzycia danej jednostki. Niniejszy artykul prezentuje analize francuskiego przymiotnika piquant i jego klasyfikacje w oparciu klasy obiektowe. Wyszczegolniając tyle znaczen danego leksemu, ile typow klas obiektowych, do ktorych tenze przymiotnik jest stosowany, mozemy odnaleźc jego odpowiedni ekwiwalent w jezyku docelowym, a w konsekwencji podac jego poprawne tlumaczenie.

Research into modality has tended to focus on modal auxiliary verbs (modals) at the expense of other forms that may express modal meaning. This thesis takes a phraseological, exploratory approach to the investigation of modal meaning by focusing on modal expressions with verbs with wh-clause complementation (the V wh pattern). The approach first tests the hypothesis that the pattern is associated with markers of modal meaning and then goes on to conduct a concordance analysis of samples of frequently-occurring V wh verbs taken from the British National Corpus. This analysis first categorizes these verbs into semantic sets and then explores which realizations of different types of modal meaning -obligation, volition, potential, and uncertainty -are most often found with verbs in particular sets. The presentation of the results of this analysis also involves a discussion of how exponents of modal meaning other than modals extend the range of expression available to users of English, indicating what an exclusive focus on modals will tend to overlook. 'have + modal noun + to-infinitive' construction.

La finalidad de este artículo es el estudio de los verbos con partícula, estableciendo una delimitación clara y precisa entre verbos adverbiales y preposicionales, ya que generalmente las partículas utilizadas son las mismas en ambos tipos de verbos. Los mecanismos o dispositivos utilizados para su diferenciación paradigmática son: movilidad de la partícula (precediendo o siguiendo al objeto), transformación temática ("cleft sentence") e inclusión de un adverbio entre el verbo y la partícula.

This paper presents the current state of ViPEr, the Lexicon-Grammar of European Portuguese verbs, a database with distributional, syntactic and semantic properties of the most frequently occurring verbs. The classification follows the theoretical framework of the Lexicon-Grammar. The paper presents the main linguistic criteria that were adopted in the classification of 5052 frequently occurring verbs, which yield 6,059 different constructions or word senses. The paper concludes with some preliminary results on the application of ViPEr to texts and plans for future work.

Deux corpus en anglais et en francais d'environ 10 000 mots chacun ont ete etiquetes manuellement et les taux d'occurrence des determinants en anglais et en francais ont ete calcules pour tous les SN (a l'exception des pronoms et des Np). Ensuite les taux d'occurrence de chaque mode de determination ont ete compares pour les SN sans epithete ni attribut, les SN sujets d'un predicat attributif mais sans epithete, les SN comportant une seule epithete de nature adjectivale, et ceux comportant des epithetes multiples et/ou de nature diverse. Une distinction a ete faite en francais entre les SN a epithete postposee et anteposee. Pour faciliter la comparaison, un petit nombre de SN cumulant plusieurs modes de qualification en meme temps (epithete + attribut) ont ete exclus. Il s'est ainsi avere que les taux d'occurrence des differents modes de determination varient sensiblement selon le type de qualification adjectivale. Si les articles definis sont toujours le...

In this paper we propose an approach for identifying syntactic behaviours related to lexical items and linking them to the meanings. This approach is based on the analysis of the textual content presented in LMF normalized dictionaries by means of Definition and Context classes. The main particularity of these contents is their large availability and their semantically control due to their attachment to the meanings, which promotes the effective links between the syntactic behaviours and the meaning. In order to test the performance of the proposed approach, we tested it on an available Arabic LMF normalized dictionary. The experiment treats 9,800 verbs and allows us to evaluate the identified syntactic behaviours as well as their links to the meanings.

This paper describes the conversion into LMF, a standard lexicographic digital format of ‘al-qāmūs al-muḥīṭ, a Medieval Arabic lexicon. The lexicon is first described, then all the steps required for the conversion are illustrated. The work is will produce a useful lexicographic resource for Arabic NLP, but is also interesting per se, to study the implications of adapting the LMF model to the Arabic language. Some reflections are offered as to the status of roots with respect to previously suggested representations. In particular, roots are, in our opinion are to be not treated as lexical entries, but modeled as lexical metadata for classifying and identifying lexical entries. In this manner, each root connects all entries that are derived from it.

La documentación dels arxius aragonesos que conserven originals redactats en català no permet, en la majoria dels casos, reumtar-se més enllà del segle XIV; per aquest motiu, tant des d'una perspectiva filològica com històrica, els documents de la baixa edat mitjana, en particular, i la resta de testimonis que es custodien en els fons aragonesos, en general, contribueixen a la recopilació de referenències essencials sobre la llengua antiga de l'Aragó catalanòfon. En la present publicació s'ofereix l'edició de 50 manuscrits en pergamí del segle XIV, en els quals són progagonistes els pobles i la gent del Matarranya (Terol), i s'hi analitza la llengua catalana reflectida en aquella scripta.

This work documents the project and development of various computational linguistic resources that support the Brazilian Portuguese language according to the formal methodology used by the corpus processing system called UNITEX. The delivered resources include computational lexicons, libraries to access compressed lexicons, and additional tools to validate those resources.

The paper describes Portuguese large-scale linguistic resources, mainly computational lexicons and grammars, developed by LabEL. These resources are formalized and applied to texts by means of finite-state techniques, more and more acknowledged in Natural Language Processing. On the one hand, it illustrates methods on lexical representation for simple words and multi-word expressions; on the other hand, it provides examples (in form of concordances) of linguistic structures recognized after the application of disambiguation and parsing grammars to texts. The paper ends with a short reference to the publicly available data highlighting its contribution towards dissemination of LabEL's knowledge on language technology.

English. The present research deals with the automatic annotation and classification of vulgar ad offensive speech on social media. In this paper we will test the effectiveness of the computational treatment of the taboo contents shared on the web, the output is a corpus of 31,749 Facebook comments which has been automatically annotated through a lexicon-based method for the automatic identification and classification of taboo expressions. Italiano. La presente ricerca affronta il tema dell'annotazione e della classificazione automatica dei contenuti volgari e offensivi espressi nei social media. Lo scopo del nostro lavoro consiste nel testare l'efficacia del trattamento computazionale dei contenuti tabù condivisi in rete. L'output che forniamo un corpus di 31,749 commenti generati dagli utenti di Facebook e annotato automaticamente attraverso un metodo basato sul lessico per l'identificazione e la classificazione delle espressioni tabù.

Based on an electronic dictionary of more than 15,000 Spanish adjectives, that we elaborated in the frame of the research laboratory fLexSem (Phonetics, Lexicology and Semantics, UAB), this paper describes all the cases where a multilexical unit in French is given as a translation equivalent for an unilexical adjective in Spanish. The aim of this description is, on the one hand, to establish the notion of polylexical adjective as a necessary category in translation equivalence for bilingual dictionaries. On the other hand, we have also as goal to provide a full formal typology of these translation equivalents..

We propose two discriminative strategies to integrate comp ound word recognition in a parsing context: (i) compound pregrouping with Condition al Random Fields before parsing, (ii) reranking parses with a maximum entropy model after par sing. These discriminative models integrate features dedicated to compounds, some of theme be ing computed from external lexical resources. We show that the pregrouping strategy largely ov ertakes current state-of-the-art results, while the reranking strategy is somewhat deceivin g. All the experiments conducted in this paper open new interesting research directions.

Based on an electronic dictionary of more than 15,000 Spanish adjectives, that we elaborated in the frame of the research laboratory fLexSem (Phonetics, Lexicology and Semantics, UAB), this paper describes all the cases where a multilexical unit in French is given as a translation equivalent for an unilexical adjective in Spanish. The aim of this description is, on the one hand, to establish the notion of polylexical adjective as a necessary category in translation equivalence for bilingual dictionaries. On the other hand, we have also as goal to provide a full formal typology of these translation equivalents..

Transducers namely transducer cascades are used in several NLP-applications such as Ara-bic named entity recognition (ANER). To ex-periment and evaluate an ANER process, a weight coverage corpus is necessary. In this pa-per, we propose an ANER method based on transducer cascade. The proposed transducer cascade is generated with the CasSys tool inte-grated in Unitex linguistic platform. The exper-imentation of our method is done on a Wikipe-dia corpus. The Wikipedia text format is ob-tained with Kiwix tool. The experiment results are satisfactory.

Certains verbes ont une double commande modale : ils admettent une construction permettant l’usage du subjonctif et de l’indicatif dans la complétive qui leur est adjointe. Ainsi, ces verbes se trouvent dans des contextes polysémiques à tous les niveaux d’analyse (lexicale, syntaxique, grammaticale et pragmatique). Dans un tel cas, le mode peut représenter une marque formelle désambiguïsante. Plus précisément, parfois le verbe régissant (dans la principale) est polysémique en lui-même, et selon son sens il commande soit l’indicatif, soit le subjonctif dans la subordonnée complétive : il s’agit de la polysémie lexicale, qui peut être forte ou faible. D’un autre point de vue, certains verbes à l’indicatif/subjonctif dans la subordonnée modifient plus ou moins légèrement le sens de la phrase entière : on considère ici cela comme un cas de polysémie syntaxique, étant donné que ces verbes apparaissent dans diverses structures et modalités.

We present an electronic dictionary of French multiword adverbs. This dictionary is designed for investigation on information retrieval and extraction, automatic lexical acquisition, as well as on deep and shallow syntactic parsing. We delimit the scope of the dictionary in terms of lexical coverage and of grammatical coverage, we outline the formal description of entries, and we give an overview of the syntactic and semantic features which are associated to the 6,800 adverbial entries of the lexicon. This electronic dictionary is freely available on the web.

This paper presents a French corpus annotated for multiword nouns. This corpus is designed for investigation in information retrieval and extraction, as well as in deep and shallow syntactic parsing. We delimit which kind of multiword units we targeted for this annotation task; we describe the resources and methods we used for the annotation; and we briefly comment on the results. The annotated corpus is available at http://infolingu.univ-mlv.fr/ under the LGPLLR license.

This paper presents a French corpus annotated for multiword expressions (MWEs) with adverbial function. This corpus is designed for investigation on information retrieval and extraction, as well as on deep and shallow syntactic parsing. We delimit which kind of MWEs we annotated, we describe the resources and methods we used for the annotation, and we briefly comment the results. The annotated corpus is available at http://infolingu.univ-mlv.fr/ under the LGPLLR license.

This paper presents a work on extending the adverbial entries of LGLex: a NLP oriented syntactic resource for French. Adverbs were extracted from the Lexicon-Grammar tables of both simple adverbs ending in -ment '-ly' (Molinier and Levrier, 2000) and compound adverbs (Gross, 1986; 1990). This work relies on the exploitation of fine-grained linguistic information provided in existing resources. Various features are encoded in both LG tables and they haven't been exploited yet. They describe the relations of deleting, permuting, intensifying and paraphrasing that associate, on the one hand, the simple and compound adverbs and, on the other hand, different types of compound adverbs. The resulting syntactic resource is manually evaluated and freely available under the LGPL-LR license.

Like common noun phrases, proper names contain ambiguous conjoined phrases that make their delimitation and classification difficult in text. This paper presents a finite-state approach to the disambiguation of Portuguese candidate proper name strings containing the coordinating conjunction e (and). In such name strings, the conjunction can denote a relation between two independent names, but it can also be part of a multiword proper name. The coordination of multiword independent names may involve ellipsis of some lexical constituents, which causes additional difficulties to proper name identification and classification.

Le but de cet article est d 'une part de preciser les conditions de l'emploi comitatif de la preposition avec en francais, et d'autre part d'expliquer la difference entre cet emploi et le jonctif et, qui expriment tous deux l'idee d'addition. Une etude comparative entre le francais et le coreen permet d'apporter un nouvel eclairage a cette question. L'hypothese defendue est que le comitatif et le jonctif, qui n'ont pas le meme fonctionnement d'un point de vue syntaxique et semantique, se distinguent aussi quant a l'idee d'addition qu'ils expriment; l'addition exprimee par le comitatif est d'ordre qualitatif, alors que l'addition exprimee par le jonctif est d'ordre quantitatif.

This paper describes the fuzzy boundaries between support verb constructions (SVC) with ter "have" and dar "give" and causative operator verb (VopC) constructions involving these same verbs, in Brazilian Portuguese (BP), which form a complex set of relations: (i) both verbs are the support verb of the same noun (SVC); (ii) dar is the standard (active-like) SVC while ter is a converse (passive-like) SVC; and (iii) dar is a VopC, operating on a ter SVC. In this paper we have systematically studied these complex relations involving SVC and VopC for BP, which constitute a challenge to Natural Language Processing (NLP) systems, and have been often ignored in related work. The paper proposes a lexically-based strategy to implement SVC in a fully-fledged, rule-based parsing system, yielding an adequate semantic structure of the events (predicates) denoted by predicative nouns in SVC.