Cyril Grouin

Laboratoire Interdisciplinaire des Sciences du Numérique

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Pierre Zweigenbaum

Université Paris-Saclay

Natalia Grabar

Savoirs, Textes, Langage

Thierry Hamon

Institut de l'Ouest : Droit et Europe

Sophie Rosset

Université Paris-Saclay

Patrick Paroubek

Centre National de la Recherche Scientifique

Louise Deléger

Mathématiques et Informatique Appliquées du Génome à l'Environnement

Aurélie Névéol

Laboratoire Interdisciplinaire des Sciences du Numérique

Thomas Lavergne

Norwegian Meteorological Institute

Vincent Claveau

Centre National de la Recherche Scientifique

Anne‐Laure Ligozat

Université Paris-Saclay

Cooperative Institutions

Centre National de la Recherche Scientifique

123

Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur

Inserm

Université Paris-Saclay

Dalle Molle Institute for Artificial Intelligence Research

East Stroudsburg University

Université d'Orléans

RMIT University

Brandeis University

Université Paris-Sud

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Actes de l'atelier DeFT - Défi Fouille de Textes

Natalia Grabar Cyril Grouin Thierry Hamon Vincent Claveau

Source

Cite

Citations (0)

Vivre avec le SDRC

Marie-France Pilarski Nadège Sieckelinck Johanne Sénécaux Christine Le Clainche Aurélia Lamy

Cite

Citations (0)

Automatic pre-annotation of clinical texts to support the dialogue with domain experts during the design of an annotation schema

Virgile Barthet Marie José Aroulanda Laura Monceaux-Cachard Christine Jacquin Cyril Grouin

Schema (genetic algorithms)

Source

Cite

Citations (0)

Automatic classification of free-text medical causes from death certificates for reactive mortality surveillance in France

International Journal of Medical Informatics (2019)

Yasmine Baghdadi Alix Bourrée Aude Robert Grégoire Rey Anne Gallay

10.1016/j.ijmedinf.2019.06.022

Cite

Citations (11)

Analyse automatique de textes littéraires et scientifiques : présentation et résultats du défi fouille de texte DEFT2014

Thierry Hamon Quentin Pleplé Patrick Paroubek Pierre Zweigenbaum Cyril Grouin

In this paper, we present the 2014 DEFT text mining shared task, dedicated to the analysis of literature texts (corpus Short Edition) and scientific texts (TALN archives) through four tasks: identifying the literary type, eval- uating writing quality, determining whether the quality of a work achieves consensus among the reviewers, and finally identifying the conference session of a scientific paper. In order to evaluate the results, we used normalized discounted cumulative gain (NDCG, task 1), accuracy of the relative distance to the mean solution (EDRM, task 2), precision (task 3), and correction (task 4). The results obtained by the participants are highly contrasted and reveal the difficulty of each task, although one system reached the maximal performance in task 4. Mots-cles : Fouille d'opinion, classification automatique, evaluation.

Source

Cite

Citations (0)

Approches à base de fréquences pour la simplification lexicale

HAL (Le Centre pour la Communication Scientifique Directe) (2013)

Anne‐Laure Ligozat Cyril Grouin Anne Garcia-Fernandez Delphine Bernhard

La simplification lexicale consiste a remplacer des mots ou des phrases par leur equivalent plus simple. Dans cet article, nous presentons trois modeles de simplification lexicale, fondes sur differents criteres qui font qu'un mot est plus simple a lire et a comprendre qu'un autre. Nous avons teste differentes tailles de contextes autour du mot etudie : absence de contexte avec un modele fonde sur des frequences de termes dans un corpus d'anglais simplifie ; quelques mots de contexte au moyen de probabilites a base de n-grammes issus de donnees du web ; et le contexte etendu avec un modele fonde sur les frequences de cooccurrences. ABSTRACT Studying frequency-based approaches to process lexical simplification Lexical simplification aims at replacing words or phrases by simpler equivalents. In this paper, we present three models for lexical simplification, focusing on the criteria that make one word simpler to read and understand than another. We tested different contexts of the considered word : no context, with a model based on word frequencies in a simplified English corpus ; a few words context, with n-grams probabilites on Web data, and an extended context, with a model based on co-occurrence frequencies. MOTS-CLES : simplification lexicale, frequence lexicale, modele de langue.

Source

Cite

Citations (2)

Testing Tactics to Localize De-Identification

Studies in health technology and informatics (2009)

Cyril Grouin Arnaud Rosier Olivier Dameron Pierre Zweigenbaum

Recent renewed interest in de-identification (also known as “anonymisation”) has led to the development of a series of systems in the United States with very good performance on challenge test sets. De-identification needs however to be tuned to the local documents and their specificities. We address here two issues raised in this context. First, tuning is generally performed by language engineers who should not have to work on identified text. We therefore perform a first gross de-identification step in the hospital. Second, to set up a de-identification system for new documents in a language different from English, here French patient reports, we tested two methods: the first attempts to adapt an existing US de-identifier for English, the second re-develops a new system which applies the same methods. The first method involved localizing patterns designed for English, which proved cumbersome and did not quickly obtain good performance. With a similar effort, the latter method obtained much better results. Evaluated on a set of 23 randomly selected texts from a corpus of 21,749 clinical texts, it obtained 83% recall and 92% precision.

Identification

10.3233/978-1-60750-044-5-735

Cite

Citations (19)

Extracting Medication Information from French Clinical Texts

Studies in health technology and informatics (2010)

Louise Deléger Cyril Grouin Pierre Zweigenbaum

Much more Natural Language Processing (NLP) work has been performed on the English language than on any other. This general observation is also true of medical NLP, although clinical language processing needs are as strong in other languages as they are in English. In specific subdomains, such as drug prescription, the expression of information can be closely related across different languages, which should help transfer systems from English to other languages. We report here the implementation of a medication extraction system which extracts drugs and related information from French clinical texts, on the basis of an approach initially designed for English within the framework of the i2b2 2009 challenge. The system relies on specialized lexicons and a set of extraction rules. A first evaluation on 50 annotated texts obtains 86.7% F-measure, a level higher than the original English system and close to related work. This shows that the same rule-based approach can be applied to English and French languages, with a similar level of performance. We further discuss directions for improving both systems.

10.3233/978-1-60750-588-4-949

Cite

Citations (21)

Recycling an Information Extraction System to Automatically Produce Semantic Annotations for the Web.

Thierry Poibeau Alexandre Arcouteil Cyril Grouin

Source

Cite

Citations (3)

Corpus annoté de cas cliniques en français (Annotated corpus with clinical cases in French)

Natalia Grabar Cyril Grouin Thierry Hamon Vincent Claveau

Les corpus textuels sont utiles pour diverses applications de traitement automatique des langues (TAL) en fournissant les donnees necessaires pour leur creation, adaptation ou evaluation. Cependant, dans certains domaines comme le domaine medical, l’acces aux donnees est rendu complique, voire impossible, pour des raisons de confidentialite et d’ethique. Il existe neanmoins de reels besoins en corpus cliniques pour l’enseignement et la recherche. Pour repondre a ce defi, nous presentons dans cet article le corpus CAS contenant des cas cliniques de patients, reels ou fictifs, que nous avons compiles. Ces cas cliniques en francais couvrent plusieurs specialites medicales et focalisent donc sur differentes situations cliniques. Actuellement, le corpus contient 4 300 cas (environ 1,5M d’occurrences de mots). Il est accompagne d’informations (discussions des cas cliniques, mots-cles, etc.) et d’annotations que nous avons effectuees au regard des besoins de la recherche en TAL dans ce domaine. Nous presentons egalement les resultats de premieres experiences de recherche et d’extraction d’information qui ont ete effectuees avec ce corpus annote. Ces experiences peuvent fournir une baseline a d’autres chercheurs souhaitant travailler avec les donnees.

Source

Cite

Citations (0)