Verbal and nominal predicate structures present interesting linguistic properties. In biological texts they express relations between biological entities. We describe PredXtract, a platform which extracts predicate argument structures, and present the results. Mots-clés : Predicate structure, Nominalizations of verbs, Dependency grammar, Link Parser.

This paper presents ZOMBILINGO, a Game With A Purpose (GWAP) that allows for the dependency syntax annotation of French corpora. The created resource is freely available on the game Web site. Mots-clés : jeux ayant un but, complexité, annotation, syntaxe en dépendances.

Résumé. Les termes-clés sont les mots ou les expressions polylexicales qui représentent le contenu principal d’un document. Ils sont utiles pour diverses applications, telles que l’indexation automatique ou le résumé automatique, mais ne sont pas toujours disponibles. De ce fait, nous nous intéressons à l’extraction automatique de termes-clés et, plus particulièrement, à la difficulté de cette ...

This papers show that new web technologies such as SVG, DOM, AJAX and CSS, are now mature enough to allow browsing of musical scores with optimal quality for the graphical and ergonomical parts, together with XML powerfull standard data-mining tools. MOTS-CLÉS : AJAX, DOM, DTD, CSS, partitions musicales, MusicXML, SAX, SVG, web.

The authorship attribution literature demonstrates the difficulty to design classifiers that outperform simple strategies such as linear classifiers operating on bag of features representation of documents. To overcome this difficulty we propose to use Bagging techniques that rely on learning classifiers on different random subsets of features, then to combine their decision by making them vote...

In this paper we describe some interactions between collocations and automatic text categorization. First, we use the differents categories to extract strings (through collocations agglutinations) related to each categorie. Then we use these categories-specific strings to improve categorization. MOTS-CLÉS : collocations, catégorisation automatique de textes.

This paper deals with how we used a “knowledge organization system” management infrastructure to build software for researchers involved in text interpretation. Founded on a viewpoints-based model, this implementation highlighted the heterogeneity of analyses structures, whether it comes from authors diversity or from automation level. MOTS-CLÉS : Analyse qualitative, annotation, humanités numé...

This paper presents a statistical approach to tree to tree correction adapted to structural information retrieval. It consits on indexing content and structure and use the both index in an information retrieval process. Retrieval process uses extended document and query structures and returns some fragments that follow the content and structure of the original query. MOTS-CLÉS : recherche d’inf...

We propose two discriminative strategies to integrate compound word recognition in a parsing context: (i) compound pregrouping with Conditional Random Fields before parsing, (ii) reranking parses with a maximum entropy model after parsing. These discriminative models integrate features dedicated to compounds, some of theme being computed from external lexical resources. We show that the pregrou...

This paper presents and evaluates an original approach to automatically align bitexts at the word level. It relies on a syntactic dependency analysis of the texts and uses a machinelearning technique, namely inductive logic programming, to automatically infer rules called propagation rules. These rules make the most of the syntactic information to precisely align words. This approach is entirel...