TEXTE: Exploration et exploitation de donnees textuelles

Nous développons des modèles et des outils pour analyse automatique, syntaxique et sémantique, du langage naturel ainsi que pour la constitution des ressources nécessaires. 

Membres

Permanents

Non permanents

Thématiques de Recherche

L’équipe TEXTE développe des méthodes, des outils et des ressources pour le traitement automatique du langage naturel, surtout écrit.  Ces travaux portent plus particulièrement sur sa syntaxe et sur sa sémantique aussi bien logique que lexicale. Nous utilisons plutôt des méthodes symboliques, le plus souvent logiques,  d'où notre rattachement au pôle Intelligence artificielle.  Bien qu'elles soient toutes reliées entre elles, distinguons dans Texte les activités suivantes: 

  • Construction, acquisition de ressources pour le traitement automatique des langues (lexique, grammaire)
  • Analyse automatique de la syntaxe et de la sémantique du langage naturel.

Ces travaux nécessitent des recherches fondamentales,  souvent fédérées par la logique:

  • Programmation logique par contraintes pour la syntaxe guidée par les modèles
  • Analyse syntaxique et  sémantique en théorie des types.
  • Règles d'inférence dans un réseau lexical.
  • Représentation des connaissances.

D'autres méthodes sont aussi utilisées: jeux sérieux collaboratifs, algorithmique distribuée sur des graphes (fourmis), algèbre linéaire (vecteurs de mots), statistiques (suppression du bruit, étiquetage grammatical).

Publications depuis 2014 - Evaluation 2019

Articles de revues internationales

2019

  1. Natural Language Semantics and Computability
    Richard Moot, Christian Retoré
    Journal of Logic, Language and Information, Springer Verlag, 2019, 28 (2), pp.287-307.

2018

  1. Combining logical and distributional methods in type-logical grammars
    Richard Moot
    Journal of Language Modelling, Institute of Computer Science, Polish Academy of Sciences, Poland, 2018, 6 (2), pp.287-317.

2017

  1. From logical and linguistic generics to Hilbert’s tau and epsilon quantifiers
    Stergios Chatzikyriakidis, Fabio Pasquali, Christian Retoré
    IfColog Journal of Logics and their Applications (FLAP), College Publications, 2017, Hilbert’s epsilon and tau in Logic, Informatics and Linguistics, 4 (2), pp.231-256.
  2. An outline of type-theoretical approaches to lexical semantics
    Robin Cooper, Christian Retoré
    Journal of Language Modelling, Institute of Computer Science, Polish Academy of Sciences, Poland, 2017, 5 (2), pp.165-178.
  3. Type Theories and Lexical Networks: Using Serious Games as the Basis for Multi-Sorted Typed Systems
    Stergios Chatzikyriakidis, Mathieu Lafourcade, Lionel Ramadier, Manel Zarrouk
    Journal of Language Modelling, Institute of Computer Science, Polish Academy of Sciences, Poland, 2017, Type-theoretical approaches to lexical semantics, 5 (2), pp.229-272.

2016

  1. Conditions d’assertion de "chaque" et de "tout" et règles de déduction du quantificateur universel
    Alda Mari, Christian Retoré
    Travaux de Linguistique : Revue Internationale de Linguistique Française, De Boeck Université, 2016, 72, pp.89-106.
  2. Quantification in Ordinary Language and Proof Theory
    Michele Abrusci, Fabio Pasquali, Christian Retoré
    Philosophia Scientiae, Editions Kime, 2016, pp.185-205.

2015

  1. Software understanding: Automatic classification of software identifiers
    Pattaraporn Warintarawej, Anne Laurent, Marianne Huchard, Mathieu Lafourcade, Pierre Pompidor
    Intelligent Data Analysis, IOS Press, 2015, 19 (4), pp.761-778.

2014

  1. From Logical to Distributional Models
    Anne Preller
    Electronic Proceedings in Theoretical Computer Science, EPTCS, 2014, 171, pp.113-131.
  2. Deverbal semantics and the Montagovian generative lexicon ΛTyn
    Livy-Maria Real-Coelho, Christian Retoré
    Journal of Logic, Language and Information, Springer Verlag, 2014, 23 (3), pp.347-366.
  3. Partially Commutative Linear Logic and Lambek Caculus with Product: Natural Deduction, Normalisation, Subformula Property
    Maxime Amblard, Christian Retoré
    IfColog Journal of Logics and their Applications (FLAP), College Publications, 2014, 1 (1), pp.53-94.
  4. Category theory, logic and formal linguistics: some connections, old and new
    Jean Gillibert, Christian Retoré
    Journal of Applied Logic, Elsevier, 2014, 12 (1), pp.1-13.
  5. Natural Language Semantics in Biproduct Dagger Categories
    Anne Preller
    Journal of Applied Logic, Elsevier, 2014, 12, pp.88-108.
  6. Inferring Relations and Annotations in Semantic Network: Application to Radiology
    Lionel Ramadier, Manel Zarrouk, Mathieu Lafourcade, Antoine Micheau
    Computación y sistemas, Instituto Politécnico Nacional IPN Centro de Investigación en Computación, 2014, 18 (3), pp.455-466.
  7. Are opinions expressed in land-use planning documents?
    Eric Kergosien, Bernard Laval, Mathieu Roche, Maguelonne Teisseire
    International Journal of Geographical Information Science, Taylor & Francis, 2014, 28 (4), pp.739-762.
  8. How to Combine Text-Mining Methods to Validate Induced Verb-Object Relations?
    Nicolas Béchet, Jacques Chauché, Violaine Prince, Mathieu Roche
    Computer Science and Information Systems, ComSIS Consortium, 2014, 11 (1), pp.133-155.
  9. How can catchy titles be generated without loss of informativeness?
    Cédric Lopez, Violaine Prince, Mathieu Roche
    Expert Systems with Applications, Elsevier, 2014, 41 (4), pp.1051-1062.

Communications internationales

2019

  1. Prediction of Missing Semantic Relations in Lexical-Semantic Network using Random Forest Classifier
    Kévin Cousot, Mehdi Mirzapour, Waleed Ragheb
    CJC PRAXILING 2019, Nov 2019, Montpellier, France.
  2. A Supervised Learning Model for the Automatic Assessment of Language Levels Based on Learner Errors
    Nicolas Ballier, Thomas Gaillat, Andrew Simpkin, Bernardo Stearns, Manon Bouyé, Manel Zarrouk
    EC-TEL 2019, EATEL, Sep 2019, Delft, Netherlands. pp.308-320.
  3. Multilingual Knowledge Base Completion by Cross-lingual Semantic Relation Inference
    Nadia Bebeshina-Clairet, Mathieu Lafourcade
    LTA: Language Technologies and Applications, Sep 2019, Leipzig, Germany. <https://fedcsis.org/2019/lta>
  4. Deductive Parsing with an Unbounded Type Lexicon
    Konstantinos Kogkalidis, Michael Moortgat, Richard Moot, Giorgos Tziafas
    SEMSPACE, Aug 2019, Riga, Latvia.
  5. Proof-Theoretic Aspects of Hybrid Type-Logical Grammars
    Symon Jory Stevens-Guille, Richard Moot, Symon Stevens-Guille
    FG: Formal Grammar, Aug 2019, Riga, Latvia. pp.84-100.
  6. Inférence des relations sémantiques dans un réseau lexico-sémantique multilingue
    Nadia Bebeshina-Clairet, Mathieu Lafourcade
    TALN: Traitement Automatique des Langues Naturelles, Jul 2019, Toulouse, France. <https://www.irit.fr/pfia2019/taln-recital/taln-recital-2019-articles-acceptes/>

2018

  1. Polysemy, Individuation and Co-Predication: a simple solution to the counting puzzle implemented using lambda-DRT and MGL
    Bruno Mery, Richard Moot, Christian Retoré
    LENLS: Logic and Engineering of Natural Language Semantics, Nov 2018, Yokohama, Japan. <http://www.is.ocha.ac.jp/~bekki/lenls/>
  2. Si beaucoup d'oiseaux volent, peut-on en déduire que tous les oiseaux volent ? Production endogène de règles déductives dans un réseau lexico-sémantique
    Mathieu Lafourcade, Alain Joubert
    dans un réseau lexico-sémantique. TALIA - PFIA 2018, Jul 2018, Nancy, France.
  3. Cheap, Fast and Good! Voting Games with a Purpose
    Karën Fort, Mathieu Lafourcade, Nathalie Le Brun
    Games4NLP: Games and Gamification for Natural Language Processing , May 2018, Miyazaki, Japan.
  4. Implicit and Explicit Aspect Extraction in Financial Microblogs
    Thomas Gaillat, Bernardo Stearns, Gopal Sridhar, Ross Mcdermott, Manel Zarrouk, Brian Davis
    ECONLP: Economics and Natural Language Processing, Jul 2018, Melbourne, Australia. pp.55-61.
  5. Categorial Proof Nets and Dependency Locality: A New Metric for Linguistic Complexity
    Mehdi Mirzapour, Jean-Philippe Prost, Christian Retoré
    LACompLing: Logic and Algorithms in Computational Linguistics, Aug 2018, Stockholm, Sweden. pp.73-86.
  6. Venn Diagram and Evaluation of Syllogisms with Negative Terms: A New Algorithm
    Mehdi Mirzapour, Christian Retoré
    Diagrams: Diagrammatic Representation and Inference, Jun 2018, Edinburgh, United Kingdom. pp.716-720.
  7. FinSentiA: Sentiment Analysis in English Financial Microblogs
    Thomas Gaillat, Annanda Sousa, Manel Zarrouk, Brian Davis
    Conférence CORIA TALN 2018, Inria Rennes Bretagne-Atlantique; IRISA, May 2018, Rennes, France. <https://project.inria.fr/coriataln2018/fr/>
  8. The SSIX Corpora: Three Gold Standard Corpora for Sentiment Analysis in English, Spanish and German Financial Microblogs
    Thomas Gaillat, Manel Zarrouk, André Freitas, Brian Davis
    LREC: Language Resources and Evaluation‎ Conference, May 2018, Miyazaki, Japan. pp.2671-2675.

2017

  1. Collecting Weighted Coercions from Crowd-Sourced Lexical Data for Compositional Semantic Analysis
    Mathieu Lafourcade, Bruno Mery, Mehdi Mirzapour, Richard Moot, Christian Retoré
    isAI: International Symposium on Artificial Intelligence, Nov 2017, Tokyo, Japan. pp.214-230.
  2. Ontolex JeuxDeMots and Its Alignment to the Linguistic Linked Open Data Cloud
    Andon Tchechmedjiev, Théophile Mandon, Mathieu Lafourcade, Anne Laurent, Konstantin Todorov
    ISWC: International Semantic Web Conference, Oct 2017, Vienne, Austria. pp.678-693.
  3. Quantifier Scoping and Semantic Preferences
    Davide Catta, Mehdi Mirzapour
    CONLI: Computing Natural Language Inference, Sep 2017, Montpellier, France. <https://sites.google.com/view/conli/home>
  4. Identifying Polysemous Words and Inferring Sense Glosses in a Semantic Network
    Maxime Chapuis, Mathieu Lafourcade
    IWCS: International Conference on Computational Semantics, Sep 2017, Montpellier, France. <https://www.lirmm.fr/iwcs2017/>
  5. Towards the Automatic Detection of Nutritional Incompatibilities Based on Recipe Titles
    Nadia Bebeshina-Clairet, Mathieu Lafourcade
    CD-MAKE: Cross-Domain Conference for Machine Learning and Knowledge Extraction, Aug 2017, Reggio, Italy. pp.346-366.
  6. Finding Missing Categories in Incomplete Utterances
    Mehdi Mirzapour
    TALN: Traitement Automatique des Langues Naturelles, Jun 2017, Orléans, France. <http://www.atala.org/content/taln-recital-2017-24ème-conférence-sur-le-traitement-automatique-des-langues-naturelles>
  7. Collecting Crowd-Sourced Lexical Coercions for Compositional Semantic Analysis
    Mathieu Lafourcade, Bruno Mery, Mehdi Mirzapour, Richard Moot, Christian Retoré
    LENLS: Logic and Engineering of Natural Language Semantics, Nov 2017, Tokyo, Japan. <http://www.is.ocha.ac.jp/~bekki/lenls/lenls14/index.html>
  8. Ambiguss, a game for building a Sense Annotated Corpus for French
    Mathieu Lafourcade, Nathalie Le Brun
    IWCS: International Conference on Computational Semantics, Sep 2017, Montpellier, France. <https://www.lirmm.fr/iwcs2017/>
  9. Explicative Path Finding in a Semantic Network
    Kévin Cousot, Mathieu Lafourcade
    CONLI: Computing Natural Language Inference, Sep 2017, Montpellier, France. <https://sites.google.com/view/conli/>
  10. If mice were reptiles, then reptiles could be mammals or How to detect errors in the JeuxDeMots lexical network?
    Mathieu Lafourcade, Alain Joubert, Nathalie Le Brun
    RANLP: Recent Advances in Natural Language Processing, Sep 2017, Varna, Bulgaria. <http://lml.bas.bg/ranlp2017/>
  11. Parcourir, reconnaître et réfléchir. Combinaison de méthodes légères pour l'extraction de relations sémantiques.
    Mathieu Lafourcade, Nathalie Le Brun
    TALN: Traitement Automatique des Langues Naturelles, Jun 2017, Orléans, France. <http://taln2017.cnrs.fr>
  12. An Empirical Study for a Machine Aided Translation of French Prepositions 'à', 'de' and 'en' into English
    Violaine Prince
    LTC: Language and Technology Conference, Nov 2017, Poznan, Poland. <http://ltc.amu.edu.pl/a2017/>

2016

  1. Compilation de grammaire de propriétés pour l'analyse syntaxique par optimisation de contraintes
    Jean-Philippe Prost, Remi Coletta, Christophe Lecoutre
    TALN: Traitement Automatique des Langues Naturelles, Jul 2016, Paris, France. <https://jep-taln2016.limsi.fr/>
  2. Mixing Crowdsourcing and Graph Propagation to Build a Sentiment Lexicon
    Mathieu Lafourcade, Nathalie Le Brun, Alain Joubert
    Feelings are contagious. NLDB: Natural Language Processing and Information Systems, Jun 2016, Manchester, United Kingdom. pp.258-266.
  3. Découverte des patrons de connaissance grâce à la modélisation sémantique des phrases d'instructions
    Nadia Bebeshina-Clairet, Sylvie Despres, Mathieu Lafourcade
    TOTh: Terminology and Ontology - Theories and applications, Jun 2016, Chambéry, France. <http://porphyre.org/toth/toth-2016>
  4. Using Constraints on a general Knowledge lexical networK for domain-specific semantic relation extraction and modeling
    Nadia Bebeshina-Clairet, Lionel Ramadier, Mathieu Lafourcade
    DIALOGUE, Jun 2016, Moscou, Russia. <http://www.dialog-21.ru/en/dialogue2016/results/>
  5. Construire un lexique de sentiments par crowdsourcing et propagation
    Mathieu Lafourcade, Nathalie Le Brun, Alain Joubert
    TALN: Traitement Automatique des Langues Naturelles, Jul 2016, Paris, France. <https://jep-taln2016.limsi.fr/actes/index.php?lang=fr>
  6. Semantic RelationExtraction with Semantic Patterns: Experiment on Radiology Report
    Mathieu Lafourcade, Lionel Ramadier
    LREC: Language Resources and Evaluation Conference, May 2016, Portorož, Slovenia. <http://lrec2016.lrec-conf.org/en/>
  7. Patrons sémantiques pour l'extraction de relations entre termes - Application aux comptes rendus radiologiques
    Lionel Ramadier, Mathieu Lafourcade
    TALN: Traitement Automatique des Langues Naturelles, Jul 2016, Paris, France. <https://jep-taln2016.limsi.fr>

2015

  1. “Chaque vin a sa lie." versus “Toute nuit a un jour." --- does the difference in the human processing of " chaque" and " tout" match the difference between the proof rules for conjunction and quantification?
    Alda Mari, Christian Retoré
    (In)Coherence of discourse, M. Amblard; M. Musiol; M. Rebuschi, Dec 2015, Nancy, France. <http://discours.loria.fr/2015/>
  2. Medical Imaging Report Indexing: Enrichment of Index through an Algorithm of Spreading over a Lexico-semantic Network
    Mathieu Lafourcade, Lionel Ramadier
    RANLP: Recent Advances in Natural Language Processing, Sep 2015, Hissar, Bulgaria. <http://lml.bas.bg/ranlp2015/start1.php>
  3. Typed Hilbert Operators for the Lexical Semantics of Singular and Plural Determiner Phrases
    Bruno Mery, Richard Moot, Christian Retoré
    Epsilon: Hilbert’s Epsilon and Tau in Logic, Informatics and Linguistics, Christian Retoré; Stergios Chatzikyriakidis; Fabio Pasquali, Aug 2015, Montpellier, France. <https://sites.google.com/site/epsilon2015workshop/>
  4. Type Theories and Lexical Networks: Using Serious Games as the Basis for Multi-Sorted Typed Systems
    Stergios Chatzikyriakidis, Mathieu Lafourcade, Lionel Ramadier, Manel Zarrouk
    ESSLLI: European Summer School in Logic, Language and Information, Aug 2015, Barcelona, Spain. <http://eventia.upf.edu/congresos/ficha.en.html?cc=146>
  5. The Origin of the Distribution Doctrine
    Mehdi Mirzapour
    Universal Logic, Jun 2015, Istanbul, Turkey. <https://www.uni-log.org/start5.html>
  6. Quantifier scope: a formal and experimental study
    Arthur Capelier-Mourguy, Philippe Blache, Christian Retoré, Laurent Prevot
    CJC-SC: Colloque des Jeunes Chercheurs en Sciences Cognitives, Jun 2015, Compiègne, France. <http://cjcsc.sciencesconf.org>
  7. A Case Study of Copredication over a Deverbal that Reconciles Empirical Data with Computational Semantics
    Livy Real, Christian Retoré
    LENLS: Logic and Engineering of Natural Language Semantics, Nov 2015, Tokyo, Japan. <http://www.is.ocha.ac.jp/~bekki/lenls/lenls12/index.html>
  8. Are Books Events? Ontological Inclusions as Coercive Sub-Typing, Lexical Transfers as Entailment
    Bruno Mery, Christian Retoré
    LENLS: Logic and Engineering of Natural Language Semantics, Nov 2015, Tokyo, Japan. pp.74-87.
  9. Vous aimez ?...ou pas ? LikeIt, un jeu pour construire une ressource lexicale de polarité.
    Mathieu Lafourcade, Nathalie Le Brun, Alain Joubert
    TALN: Traitement Automatique des Langues Naturelles, Jun 2015, Caen, France. <http://www.atala.org/taln_archives/TALN/TALN-2015/>

2014

  1. Computing the Semantics of Plurals and Massive Entities Using Many-Sorted Types
    Bruno Mery, Richard Moot, Christian Retoré
    LENLS: Logic and Engineering of Natural Language Semantics, JSAI / isAI, Nov 2014, Kanagawa, Japan. pp.144-159.
  2. From NL Preference Expressions to Comparative Preference Statements: A Preliminary Study in Eliciting Preferences for Customised Decision Support.
    Souhila Kaci, Namrata Patel, Violaine Prince
    ICTAI: International Conference on Tools with Artificial Intelligence, Nov 2014, Limassol, Cyprus. pp.591-598.
  3. Mining Tweet Data - Statistic and semantic information for political tweet classification
    Guillaume Tisserant, Mathieu Roche, Violaine Prince
    KDIR: Knowledge Discovery and Information Retrieval, Oct 2014, Rome, Italy. pp.523-529.
  4. Typed Hilbert Epsilon Operators and the Semantics of Determiner Phrases
    Christian Retoré
    FG: Formal Grammar, Aug 2014, Tübingen, Germany. pp.15-33.
  5. Les couleurs des gens
    Mathieu Lafourcade, Nathalie Le Brun, Virginie Zampa
    TALN: Traitement Automatique des Langues Naturelles, Jul 2014, Marseille, France. <https://www.atala.org/TALN-RECITAL-2014-21eme-conference>
  6. Crowdsourcing Word-Color Associations
    Mathieu Lafourcade, Nathalie Le Brun, Virginie Zampa
    NLDB: Natural Language Processing and Information Systems, Jun 2014, Montpellier, France. pp.39-44.
  7. Propa-L: a Semantic Filtering Service from a Lexical Network Created using Games With A Purpose
    Mathieu Lafourcade, Karen Fort
    LREC: Language Resources and Evaluation Conference, May 2014, Reykjavik, Iceland. pp.1676-1681.
  8. From Natural Language to RDF Graphs with Pregroups
    Antonin Delpeuch, Anne Preller
    TTNLS: Type Theory and Natural Language Semantics, Apr 2014, Gothenburg, Sweden. pp.55-62.
  9. Vectorisation paramétrée des données textuelles
    Célia da Costa Pereira, Mathieu Lafourcade, Patrick Lloret, Cédric Lopez, Mathieu Roche
    EGC: Extraction et Gestion des Connaissances, Jan 2014, Rennes, France. pp.593-596.
  10. Spreading Relation Annotations in a Lexical Semantic Network Applied to Radiology
    Lionel Ramadier, Manel Zarrouk, Mathieu Lafourcade, Antoine Micheau
    CICLing: Computational Linguistics and Intelligent Text Processing, Apr 2014, Kathmandu, Nepal. pp.40-51.
  11. Jugement exact de grammaticalité d'arbre syntaxique probable
    Jean-Philippe Prost
    TALN: Traitement Automatique des Langues Naturelles, Jul 2014, Marseille, France. <http://www.atala.org/taln_archives/TALN/TALN-2014/>

Mots-clés

Syntaxe, sémantique lexicale, sémantique compositionnelle. Constitution de ressources linguistiques. Analyse automatique.

Dernière mise à jour le 07/01/2019