Mathieu Lafourcade

Mathieu Lafourcade
Université de Montpellier | UM1 · Laboratory of Informatics, Robotics, and Microelectronics

PhD HDR

About

190
Publications
27,959
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
3,244
Citations
Additional affiliations
September 1997 - present
Université de Montpellier
Position
  • Dr
September 1997 - present
Université de Montpellier
Position
  • dr
September 1997 - present
Université de Montpellier
Position
  • Teacher - Researcher

Publications

Publications (190)
Chapter
Aujourd’hui le crowdsourcing est partout, tout en restant paradoxalement assez méconnu. Chacun d’entre nous peut ainsi être amené, parfois sans le savoir, à crowdsourcer, c’est-à-dire à produire de la donnée en tant que personne lambda, ou plus simplement à participer à des collectes et mises en forme de données. L’un des exemples les plus célèbres...
Conference Paper
Full-text available
In 2019, about 293 billion emails were sent worldwide every day. They are a valuable source of information and knowledge for professionals. Since the 90's, many studies have been done on emails and have highlighted the need for resources regarding numerous NLP tasks. Due to the lack of available resources for French, few studies on emails have been...
Article
The JeuxDeMots project aims to build a broad knowledge base in French, both common sense and specialised, using games, contributory approaches, and inference mechanisms. A dozen games have been designed as part of this project, each one allowing collecting specific information, or to consolidate the information acquired through the other games. Thi...
Conference Paper
Full-text available
GWAP design might have a tremendous effect on its popularity of course but also on the quality of the data collected. In this paper, a comparison is undertaken between two GWAPs for building term association lists, namely JeuxDeMots and Quicky Goose. After comparing both game designs, the Cohen kappa of associative lists in various configurations i...
Conference Paper
Full-text available
The task of coreference resolution applied to written texts consists in finding all lexical parts that refer to the same real world entities, properties or situations. In the aim of extracting knowledge from unstructured textual data, this task plays an essential role within the typical natural language processing pipeline. The efficiency of automa...
Chapter
In the medical domain, text simplification is both a desirable and a challenging natural language processing task. Indeed, first, medical texts can be difficult to understand for patient, because of the presence of specialized medical terms. Replacing these difficult terms with easier words can lead to improve patient’s understanding. In this paper...
Chapter
Type-theoretic frameworks for compositional semantics are aimed at producing structured meaning representations of natural language utterances.
Thesis
Full-text available
Plusieurs recherches placent le raisonnement analogique au centre du fonctionnement cognitif humain. La relation d'analogie opère entre deux paires de termes représentant deux domaines différents. Elle transfère la connaissance d'un concept connu vers un autre qu'on souhaiterait clarifier ou définir. Dans ce document, nous adressons d'abord les pro...
Conference Paper
Full-text available
We present in this paper the voting games with a purpose that were developed around JeuxDeMots, a central game aiming at creating a lexical network for French. We show that such lightweight applications can help collect quality language resources very efficiently and we advocate for a common platform for such voting games for language resources.
Conference Paper
Full-text available
Here we present the assessment of 10 years of experience concerning the JDM project, a set of GWAPs for NLP, among which a main game combined with many satellite games aims to build a large lexical-semantic network for the French language. We highlight the lessons learned from this experience for creating lexical resources through a never ending pr...
Article
Full-text available
In this paper, we show how a rich lexico-semantic network which has been built using serious games, JeuxDeMots, can help us in grounding our semantic ontologies in doing formal semantics using rich or mod- ern type theories (type theories within the tradition of Martin Löf). We discuss the issue of base types, adjectival and verbal types, hyperonym...
Conference Paper
JeuxDeMots (JdM) is a rich collaborative lexical network in French, built on a crowdsourcing principle as a game with a purpose, represented in an ad-hoc tabular format. In the interest of reuse and interoperability, we propose a conversion algorithm for JdM following the Ontolex model, along with a word sense alignment algorithm, called JdMBabeliz...
Conference Paper
Full-text available
This paper presents Ambiguss, a Game With A Purpose designed both to collect ambiguous sentences and build a Sense Annotated Corpus. It also generates a lexicon of polysemous words associated with the glosses that illustrate the different meanings. Early evaluations indicate that the approach is relevant and efficient.
Conference Paper
Full-text available
Correcting errors in a data set is a critical issue. This task can be either hand-made by experts, or by crowdsourcing methods or automatically done using algorithms. Although even if the rate of errors present in a given lexical network is rather low, it is important to reduce it. We present here automatic methods for detecting potential secondary...
Conference Paper
Full-text available
Extracting semantic relations from texts is a good way to build and supply a knowledge base, an indispensable resource for text analysis. We propose and evaluate the combination of three ways of producing lexical-semantic relations.
Conference Paper
The present paper reports experimental work on the automatic detection of nutritional incompatibilities of cooking recipes based on their titles. Such incompatibilities viewed as medical or cultural issues became a major concern in western societies. The gastronomy language represents an important challenge because of its elusiveness, its metaphors...
Conference Paper
Full-text available
Extracting semantic relations from texts is a good way to build and supply a knowledge base, an indispensable resource for text analysis. We propose and evaluate the combination of three ways of producing lexical-semantic relations.
Conference Paper
Full-text available
Correcting errors in a data set is a critical issue. This task can be either hand-made by experts, or by crowdsourcing methods, or automatically done using algorithms. We present here automatic methods for detecting potential "secondary" errors that would result from automatic inference mechanisms when they rely on an "initial" error manually detec...
Article
Full-text available
Type-theoretic frameworks for compositional semantics are aimed at producing structured meaning representations of natural language utterances. Using elements of lexical semantics, these frameworks are able to represent many difficult phenomena related to the polysemy of words and their context-dependent meanings. However, they are just as powerful...
Conference Paper
Full-text available
Cet article présente une méthode de construction d'une ressource lexicale de sentiments/émotions. Son originalité est d'associer le crowdsourcing via un GWAP (Game With A Purpose) à un algorithme de propagation, les deux ayant pour support et source de données le réseau lexical JeuxDeMots. Nous décrivons le jeu permettant de collecter des informati...
Conference Paper
Full-text available
This paper describes a method for building a sentiment lexicon. Its originality is to combine crowdsourcing via a Game With A Purpose (GWAP) with automated propagation of sentiments through a spreading algorithm, both using the lexical JeuxDeMots network as data source and substratum. We present the game designed to collect sentiment data, and the...
Article
Full-text available
In this paper, we show how a rich lexico-semantic networkwhich has been built using serious games, JeuxDeMots, can help us in grounding our semantic ontologies in doing formal semantics using rich or modern type theories (type theories within the tradition of Martin Löf). We discuss the issue of base types, adjectival and verbal types, hyperonymy/h...
Chapter
The JDM lexical network has been built thanks to on-line games the main of which, JeuxDeMots (JDM), was launched in 2007. It is currently a large lexical network, in constant evolution, containing more than 310,000 terms connected by more than 6.5 million relations. The riddle game Totaki (Tip Of the Tongue with Automated Knowledge Inferences), the...
Conference Paper
Full-text available
In medical imaging domain, digitized data is rapidly expanding Therefore it is of major interest for radiologists to be able to do an efficient and accurate extraction of imaging and clinical data (radiology reports) which are essential for a rigorous diagnosis and for a better management of patients. In daily practice, radiology reports are writte...
Conference Paper
Full-text available
Sentiment analysis from a text requires amongst others having a polarity lexical resource. We designed LikeIt, a GWAP (Game With A Purpose) that allows to attribute a positive, negative or neutral value to a term, and thus obtain a resulting polarity for most of the terms of the freely available lexical network of the JeuxDeMots project. We present...
Chapter
When it comes to resort to crowdsourcing through gaming, the medical sector is very active, whether at the research and the development of innovative processing techniques level, or that of support to diagnosis. NANODOC concerns nanomedicine and aims at the design and development of nanoparticle-based treatments. From the beginning of the game, the...
Chapter
This chapter presents the main Games With A Purpose (GWAPs) currently available in biology and biochemistry. NANOCRAFTER approaches the theme of the synthesis of desoxyribonucleic acid (DNA) fragments by using the pairing properties of nucleotides forming DNA, and especially the displacement mechanisms of strands that occur spontaneously when sever...
Chapter
Games With A Purpose (GWAPs) constitute a way to address the collaborative creation of resources, where Internauts contribute to the creation of a resource while playing, usually even without suspecting it. This chapter presents the delicate question of their acquisition after explaining the need for lexical resources in natural language processing...
Chapter
This chapter shows the great diversity of studies and/or research areas likely to give rise to games with a purpose (GWAP). The diversity comes with a relative disparity regarding the popularity of the games, which itself directly reflects the interest they generate through the mass of potential contributors that the general public represents. To c...
Chapter
The objective of the JEUXDEMOTS (JDM) project is to build a large lexical-semantic network for the French language. It involves a collection of games and counter-games that together contribute in one way or another to this objective. This chapter discusses the limitations and biases of each of the games of the project and shows how counter-games ca...
Article
Full-text available
Identifier names (e.g., packages, classes, methods, variables) are one of most important software comprehension sources. Identifier names need to be analyzed in order to support collaborative software engineering and to reuse source codes. Indeed, they convey domain concept of softwares. For instance, "getMinimumSupport" would be associated with as...
Book
Human brains can be seen as knowledge processors in a distributed system. Each of them can achieve, conscious or not, a small part of a treatment too important to be done by one. These are also "hunter / gatherers" of knowledge. Provided that the number of contributors is large enough, the results are usually better quality than if they were the re...
Conference Paper
Full-text available
Do you like it? or not ? LikeIt, a game to build a polarity lexical resource The ability to analyze the feelings that emerge from a text requires having a polarity lexical resource. In the lexical network JeuxDeMots we designed LikeIt, a GWAP that allows attributing a positive, negative or neutral value to a term, and thus obtaining for each term a...
Conference Paper
Full-text available
Les données médicales étant de plus en plus informatisées, le traitement sémantiquement efficace des rapports médicaux est devenu une nécessité. La recherche d'images radiologiques peut être grandement facilitée grâce à l'indexation textuelle des comptes rendus associés. Nous présentons un algorithme d'augmentation d'index de comptes rendus fondé s...
Book
Les jeux avec but, ou GWAP, permettent de collecter des données ou de résoudre des problèmes trop complexes, ou trop coûteux en termes de moyens pour être résolus par des machines. Ces activités ludiques, qui représentent un type de jeu sérieux, sont délicates à concevoir, car elles doivent être à la fois attrayantes et utiles. Jeux et intelligenc...
Book
Les jeux avec but, ou GWAP, permettent de collecter des données ou de résoudre des problèmes trop complexes, ou trop coûteux en termes de moyens pour être résolus par des machines. Ces activités ludiques, qui représentent un type de jeu sérieux, sont délicates à concevoir, car elles doivent être à la fois attrayantes et utiles.Jeux et intelligence...
Conference Paper
Full-text available
Domain-specific ontologies are invaluable despite many challenges associated with their development. In most cases, domain knowledge bases are built with very limited scope without considering the benefits of plunging domain knowledge to a general ontology. Furthermore, most existing resources lack meta-information about association strength (weigh...
Conference Paper
Full-text available
Automatically inferring new relations from already existing ones is a way to improve the qual-ity and coverage of a lexical network and to perform error detection. In this paper, we devise such an approach for the crowdsourced JeuxDeMots lexical network and we focus especially on word refinements. We first present deduction (generic to specific) an...
Conference Paper
Full-text available
In Natural Language Processing and semantic analysis in particular, color information may be important in order to properly process textual information (word sense disambiguation, and indexing). More specifically, knowing which colors are generally associated to terms is a crucial information. In this paper, we explore how crowdsourcing through a g...
Conference Paper
Full-text available
RÉSUMÉ Les ontologies spécifiques à un domaine ont une valeur inestimable malgré les nombreux défis liés à leur développement. Dans la plupart des cas, les bases de connaissances spécifiques à un domaine sont construites avec une portée limitée. En effet, elles ne prennent pas en compte les avantages qu'il pourrait y avoir à combiner une ontologie...
Conference Paper
Full-text available
In Natural Language Processing and semantic analysis in particular, color information may be important in order to properly process textual information (word sense disambiguation, and indexing). More specifically, knowing what colors are generally associated with what terms is crucial information. In this paper, we explore how crowdsourcing through...
Conference Paper
Full-text available
La construction d'un réseau lexico-sémantique, ainsi que sa validation, sont des tâches primordiales en Traitement Automatique des Langues. Il est également possible de densifier un tel réseau de manière endogène, grâce à des processus de raisonnement sur les données qu'il contient. Dans cet article, nous présentons un moteur d'inférences dont le b...
Article
Full-text available
This article presents Propa-L, a freely accessible Web service that allows to semantically filter a lexical network. The language resources behind the service are dynamic and created through Games With A Purpose. We show an example of application of this service: the generation of a list of keywords for parental filtering on the Web, but many other...
Conference Paper
Full-text available
Improving lexical network's quality is an important issue in the creation process of these language resources. This can be done by automatically inferring new relations from already existing ones with the purpose of (1) densifying the relations to cover the eventual lack of information and (2) detecting errors. In this paper, we devise such an appr...
Conference Paper
Full-text available
Automatically inferring new relations from already existing ones is a way to improve the quality of a lexical network by relation densification and error detection. In this paper, we devise such an approach for the JeuxDeMots lexical network, which is a freely avalaible lexical network for French. We first present deduction (generic to specific) an...
Conference Paper
Full-text available
Domain specific ontologies are invaluable but their development fac-es many challenges. In most cases, domain knowledge bases are built with very limited scope without considering the benefits of including domain knowledge to a general ontology. Furthermore, most existing resources lack meta-information about association strength (weights) and anno...
Article
Full-text available
Isolation of Vibrio cholerae O1 is necessary for cholera outbreak confirmation. Rapid diagnostic testing of fecal specimens, based on lipopolysaccharide detection of V. cholerae O1 or O139, may assist in early outbreak detection and surveillance. Cary-Blair transport medium is recommended for specimen transport. Filter paper, although used in epide...
Conference Paper
Full-text available
In Computational Linguistics, building lexical-semantic networks and validating contained relations are paramount issues as well as adding some reasoning skills in order to enrich these knowledge bases. In this paper we devise an inference engine which aims at producing new "potential" relations from already existing ones in the JeuxDeMots network....
Conference Paper
Full-text available
RESUME _______________________________________________________________________ Les réseaux lexico-sémantiques sont un type de ressources majeur en TAL. Indépendamment des stratégies de construction utilisées, inférer automatiquement des règles avec lesquelles de nouvelles relations peuvent être produites est une approche possible pour améliorer la...
Conference Paper
Full-text available
RÉSUMÉ La construction et la validation des réseaux lexico-sémantiques est un enjeu majeur en TAL. Indépendamment des stratégies de construction utilisées, inférer automatiquement de nouvelles relations à partir de celles déjà existantes est une approche possible pour améliorer la couverture et la qualité globale de la ressource. Dans ce contexte,...
Article
Full-text available
Lexical-semantic network construction and validation is a major issue in NLP. No matter the construction strategies used, automatically inferring new relations from already existing ones is a way to improve the global quality of the resource by densifying the network. In this context, the purpose of an inference engine is to formulate new conclusio...
Conference Paper
Full-text available
Lexical-semantic network construction and validation is a major issue in the NLP. No matter the construction strategies used, automatically inferring new relations from already existing ones is a way to improve the global quality of the resource by densifying the network. In this context, an inference engine has for purpose to formulate new conclus...
Chapter
Full-text available
One of the more novel approaches to collaboratively creating language resources in recent years is to use online games to collect and validate data. The most significant challenges collaborative systems face are how to train users with the necessary expertise and how to encourage participation on a scale required to produce high quality data compar...
Chapter
Les ressources lexicales (dictionnaires, bases de données, thesaurus, etc.) rassemblent des connaissances sur les mots, leurs sens et leurs usages. Si pendant des siècles elles ont été tributaires de l'imprimerie et du format textuel, il existe de nos jours une grande variété d'outils et de ressources accessibles sous des formats électroniques dive...
Article
Full-text available
Defining the drug-induced neuroadaptations specifically associated with the behavioral manifestation of addiction is a daunting task. To address this issue, we used a behavioral model that differentiates rats controlling their drug use (Non-Addict-like) from rats undergoing transition to addiction (Addict-like). Dysfunctions in prefrontal cortex (P...
Conference Paper
Full-text available
Since September 2007, a large scale lexical network for French is under construction with methods based on popular consensus by means of games (under the JeuxDeMots project). To assess the quality of such a resource built by non-expert users (players of the games), we decided to adopt an approach similar to its construction, that is to say an evalu...
Article
Full-text available
Since September 2007, a large scale lexical network for French is under construction through methods based on some kind of popular consensus by means of games (JeuxDeMots project). Human intervention can be considered as marginal. It is limited to corrections, adjustments and validation of the senses of terms, which amounts to less than 0,5 % of th...
Thesis
Full-text available
The semantic analysis of texts requires beforehand the building of objects related to lexical semantics. Idea vectors and lexical networks seems to be adequate for such a purpose and are complementary. However, one should still be able to construct them in practice. Vectors can be computed with definition corpora extracted from dictionaries, with t...
Article
Full-text available
The semantic analysis of texts requires beforehand the building of objects related to lexical semantics. Idea vectors and lexical networks seems to be adequate for such a purpose and are complementary. However, one should still be able to construct them in practice. Vectors can be computed with definition corpora extracted from dictionaries, with t...
Article
Full-text available
The reason why neurons synthesize more than one endocannabinoid (eCB) and how this is involved in the regulation of synaptic plasticity in a single neuron is not known. We found that 2-arachidonoylglycerol (2-AG) and anandamide mediate different forms of plasticity in the extended amygdala of rats. Dendritic L-type Ca(2+) channels and the subsequen...
Article
Full-text available
In this paper we investigate the possibility of a syntax– semantics inferface between a framework for Model-Theoretic Syntax on one hand and a semantic network on the other hand. We focus on exploring the ability of such a pairing to solve a collection of grammar checking problems, with an emphasis on cases of missing words. We dis-cuss a solution...
Conference Paper
Full-text available
Since September 2007, a large scale lexical network for French is under construction through methods based on some kind of popular consensus by means of games (JeuxDeMots project). Human intervention can be considered as marginal. It is limited to corrections, adjustments and validation of the senses of terms, which amounts to less than 0,5 % of th...