Search

International audience

This is the sixth publication in the series of biennial international conferences, The Future of Information Sciences (INFuture) organised by the Department of Information and Communication Sciences, Faculty of Humanities and Social Sciences, University of Zagreb. This year it is co-organised with the Miroslav Krleža Institute of Lexicography. The title of the conference is INFuture2017: Integrating ICT in Society. The conference explores the influence the information and communication sciences have on the society as a whole.

EC3 is intended to extract relevant information from large heterogeneous and multilingual text data, in particular in WEB 3.0. The project is based on an original method: contextual exploration. EC3 does not need syntactic analysis, statistical analysis or a "general" ontology. EC3 uses only small ontologies called "linguistic ontologies" that express the linguistic knowledge of a user who must concentrate on the relevant information from one point of view. This is why EC3 works very quickly on large corpus, whose components can be both whole books as well as "short texts": SMS to books. At the output, EC3 offers a visual representation of information using an original approach: the "Memory Islands". EC3 is implanted in the ACASA / LIP6 team. EC3 is tested on the large digitized corpus provided by the Labex OBVIL «Observatoire de la Vie Litteraire», in partnership with the Bibliotheque Nationale de France (http://obvil.paris-sorbonne.fr/). OBVIL intends to develop all the resources offered by digitization and computer applications to examine French literature from the sixteenth to the twentieth century, English and American literature, Italian literature, Spanish literature, in its most traditional formats and media Or the most innovative.

Using the full-text corpus of more than 75,000 research articles published by seven PLOS journals, this paper proposes a natural language processing approach for identifying the function of citations. Citation contexts are assigned based on the frequency of n-gram co-occurrences located near the citations. Results show that the most frequent linguistic patterns found in the citation contexts of papers vary according to their location in the IMRaD structure of scientific articles. The presence of negative citations is also dependent on this structure. This methodology offers new perspectives to locate these discursive forms according to the rhetorical structure of scientific articles, and will lead to a better understanding of the use of citations in scientific articles.

Scientific abstracts contain what is considered by the author(s) as information that best describe documents' content. They represent a compressed view of the informational content of a document and allow readers to evaluate the relevance of the document to a particular information need. However, little is known on their composition. This paper contributes to the understanding of the structure of abstracts, by comparing similarity between scientific abstracts and the text content of research articles. More specifically, using sentence-based similarity metrics, we quantify the phenomenon of text re-use in abstracts and examine the positions of the sentences that are similar to sentences in abstracts in the IMRaD structure (Introduction, Methods, Results and Discussion), using a corpus of over 85,000 research articles published in the seven PLOS journals. We provide evidence that 84% of abstract have at least one sentence in common with the body of the article. Our results also show that the sections of the paper from which abstract sentence are taken are invariant across the PLOS journals, with sentences mainly coming from the beginning of the introduction and the end of the conclusion., Comment: Preprint: Journal of Documentation, vol. 72, issue 4. Submitted 12-Sep-2015, accepted 10-Feb-2016. 13 pages

International audience; In this paper we consider sentences that contain Multiple In-text References (MIR) and their position in the rhetorical structure of articles. We carry out the analysis of MIR in a large scale dataset of about 80,000 research articles published by the Public Library of Science in 7 journals. We analyze two major characteristics of MIR: their positions in the IMRaD structure of articles, and the number of in-text references that make up a MIR in the different journals. We show that MIR are rather frequent in all sections of the rhetorical structure. In the Introduction section, sentences containing MIR account for more than half of the sentences with references.

Depuis l’entree en vigueur en mai 2018 du Reglement general sur la protection des donnees (RGPD), la protection des donnees personnelles est une obligation pour toutes les entreprises de l’Union europeenne. Comment faire face a cet enjeu majeur, difficile a traiter ? Plusieurs mesures techniques sont prevues, en particulier l’anonymisation et la pseudonymisation.

International audience

In this paper we study the properties of recurring in-text references in research articles and more specifically their positions in the rhetorical structure of articles and their age with respect to the citing article. We have processed a large scale corpus of approximately 80,000 papers published by PLOS (Public Library of Science). We examine the number and types of recurring in-text references, as well as their age according to positions in the IMRaD structure. (Introduction, Methods, Results, and Discussion). The results show that the age of recurring references varies considerably in all sections and journals. While they are especially dense in the Introduction section, most of them reappear in the beginning of the Results and the Discussion sections. We also observe that the beginning of the Methods and the Results sections share a significant number of recurring references.

International audience

International audience; Scientific papers are routinely structured in sections for introduction, methods, research and discussion, a standard since the 1970s. Citations originating within each section serve different purposes and can be meaningfully classified according to position, shedding light on an author's purpose for the citation. Furthermore, words near the citations in the various sections differ, providing the basis for lexical and semantic analysis of citation contexts. Approximately 50,000 scientific papers from seven PLOS journals published between 2009 and 2012 were analyzed for citation use within the identifiable document structure and for verbs used in the context of the citations. Frequencies of verbs in the four section types demonstrate the predominant use of certain words by section. Introduction sections showed greater variety of verbs, while a more limited range of verbs was seen in Methods sections. The lexical distribution process may be applied to other contexts supporting text processing based on XML format.

International audience; Nous proposons une méthodologie de fouille textuelle orientée autour des actes de citation présents dans les articles scientifiques. En étudiant la structure IMRaD ainsi que la distribution des références bibliographiques, nous pouvons mieux appréhender la structure cognitive des publications. L'analyse automatique d'un corpus de 45000 articles nous amène à dégager des propriétés intrinsèques et invariantes aux articles scientifiques. Notre méthode repose sur : a) l'identification automatique des références, b) l'extraction des contextes de citation, c) l'analyse des relations entre auteurs correspondantes aux actes de citation, et d) l'annotation sémantique de ces relations. Cette approche fait appel à une organisation de formes discursives présentes dans les contextes de citation. Ces dernières, organisées sous forme d'ontologie linguistique, serviront de ressources aux outils d'annotation sémantique. Les perspectives de cette approche se trouvent notamment dans des applications en Recherche d'Information pour l’étude de réseaux complexes en proposant des outils de filtrage sémantique autour des actes de citation. Notre prototype du système propose des interfaces de recherche et de navigation textuelle offrant de nouveaux accès plein texte aux contenus scientifiques.

The MapsMaps serve as navigational tools, documenting the landscape, warning of hazards, and highlighting potential routes of travel. Science maps chart the more abstract spaces of data and knowledge, helping forecast new fields of inquiry. Individually and as a whole, the science maps in Places & Spaces: Mapping Science use data to tell meaningful stories that both the scientist and the layperson can understand and appreciate.Every year for a decade 10 new maps were added to the Places & Spaces: Mapping Science exhibit, resulting in the 100 maps found here.; International audience; This map by computational linguists Marc Bertin and Iana Atanassova, information scientist Vincent Larivière, and historian of science Yves Gingras looks at verb occurrences found in the citation contexts of the different sections of scientific papers. Using about 50,000 scientific papers published in the PLoS series of journals over the period 2009-2012, they analyzed the full-text citation contexts to extract verbs in the introduction, methods, results, and discussion sections. Each of the four sections tends to make use of different sets of verbs. The circle in the middle was generated using the Circos tool and is meant to support the exploration of relationships between frequently used verbs and their positions in different paper sections. The linguistic contexts in which verbs occur were extracted using the Stanford part-of-speech tagger. The right side of the map shows the linguistic citation context for two verbs. On the whole, this map provides insight on the internal organization of scientific papers through the specific verbs associated with citations appearing in the different sections.

International audience; We propose a hybrid method for the extraction and characterization of citations in scientic papers using machine learning combined with rule-based approaches. Our protocol consists of the extraction of metadata, bibliography parsing, section titles processing, and nd-grained semantic annotation on the sentece level of texts. This allows us to generate Linked Open Data from a set of research papers in XML.

International audience; Information Retriveal in scientic publications can be enhanced through semantic annotation in view to identify relevant sentences containing specic semantic relations such as the expression of methods, denitions, hypotheses. We present an IR System for scientic publications that provides the possibility to lter results according to semantic facets. The semantic annotations are obtained using a rule-based method that identies specic linguistic clues organized into a linguistic ontology. This approach is implemented with Solr Search Server and offers ecient search and navigation in scientic papers. The demontrator is available at http://sempub2014.nlp-labs.org/task3/.

The last several years have seen the emergence of digital libraries from which documents are harvested using the OAI-PMH protocol. Considering the volume of data provided by these repositories, we are interested in the exploitation of the full text content of scientific publications. Our aim is to bring new value to scientific publications by automatic extraction and semantic analysis. The identification of bibliographic references in texts makes it possible to localize specific text segments that carry linguistic markers in order to annotate a set of semantic categories related to citations. This work uses a categorization of surface linguistic markers organized in a linguistic ontology. The semantic annotations are used to enrich the document metadata and to provide new types of visualizations in an information retrieval context. We present the system architecture as well as some experimental results.

International audience

International audience

National audience; Nous avons élaboré un système de recherche d'information sémantique en utilisant les annotations automatiques issues d'une ontologie discursive. Le système permet d'effectuer des recherches selon différentes catégories sémantiques liées aux citations bibliographiques dans des publications scientifiques. Nous avons proposé de nouveaux critères d'ordonnancement qui exploitent les annotations et la structure de l'ontologie. Ces critères prennent en compte la spécificité et la diversité des annotations, ainsi que les types de marqueurs linguistiques. Nous avons obtenu un score de pertinence que nous avons évalué en comparant les résultats du système avec des jugements de pertinence humains. L'évaluation montre que les algorithmes proposés produisent un ordonnancement de qualité pour les résultats se trouvant en début de liste.

International audience; Online publication of Phd theses gives young researchers a larger visibility for their work. The effective exploitation of these documents of hundreds of pages needs the development of appropriate filtering and information retrieval tools. Those currently available allow only hypertext navigation based on the table of contents or, at best, keyword search. By using the Contextual Exploration method, which identifies indicators andsemantic clues present in the context of indicators, allowing semantic annotation according to different points of view and indexing of the annotated text segments, we propose more sophisticated filtering and information retrieval, in order to obtain a quick and structured overview of the contents of a thesis.; La diffusion sur internet des thèses offre aux jeunes docteurs une valorisation plus ambitieuse et plus large de leur recherche. Pour mettre en valeur ces documents de plusieurs centaines de pages, des outils de filtrage et de recherche d'information puissants et adaptés doivent être proposés. Actuellement, les outils proposés ne permettent qu'une navigation hypertextuelle à partir de la table des matières ou, au mieux, une recherche par mots­-clés. Avec l'exploration contextuelle, méthode qui, à partir de l'identification d'un indicateur sémantique et d'indices présents dans le contexte de l'indicateur, permet l'annotation sémantique selon différents points de vue et une indexation des segments textuels annotés, nous proposons un filtrage et une recherche d'information plus élaborés permettant un aperçu rapide mais structuré du contenu de la thèse.

International audience; In this paper we present the implementation of definition extraction from multilingual corpora of scientific articles. We establish relations between the definitions and authors by using indexed references in the text. Our method is based on a linguistic ontology designed for this purpose. We propose two evaluations of the annotations.

International audience; The workshop "Mining Scientific Papers: Computational Linguistics and Bibliometrics" (CLBib 2015), co-located with the 15th International Society of Scientometrics and Informetrics Conference (ISSI 2015), brought together researchers in Bibliometrics and Computational Linguistics in order to study the ways Bibliometrics can benefit from large-scale text analytics and sense mining of scientific papers, thus exploring the interdisciplinarity of Bibliometrics and Natural Language Processing (NLP). The goals of the workshop were to answer questions like: How can we enhance author network analysis and Bibliometrics using data obtained by text analytics? What insights can NLP provide on the structure of scientific writing, on citation networks, and on in-text citation analysis? This workshop is the first step to foster the reflection on the interdisciplinarity and the benefits that the two disciplines Bibliometrics and Natural Language Processing can drive from it.

International audience

International audience; This article focuses on a Natural Language Processing (NLP) approach for the analysis of citation functions in scientific papers. Bibliometricstudies traditionally rely on citation metadata and count the number of times a publication has been cited.

International audience; Nous nous intéressons à la représentation des informations géographiques extraites d'articles scientifiques. En utilisant des outils de Traitement Automatique des Langues et de géo-codage, nous avons traité la revue PLOS Neglected Tropical Diseases afin de produire des données spatiales liées aux articles sous forme de Linked Data. Les résultats montrent une exploitation spatiale et temporelle des données représentant les zones du monde concernées par les maladies tropicales à différentes échelles.

International audience

International audience; The latest advances in research in the field of bibliometrics take into consideration in-text references. The purpose of this paper is to focus on citation contexts that express negational citations and which are relatively rare in articles. Our goal is to automate the extraction of negational citation contexts and to put them in relation to positions in the text progression of articles. After extracting sentences from the full text body of articles, we construct linguistic resources in order to identify some of the negational citation discursive patterns. We show the distribution of negational citations in the IMRaD structure. The identification of negational citations has numerous potential applications and might be used to improve information retrieval of scientific papers.

International audience; Comment traiter les langues naturelles par des systèmes de performance autres que le cerveau humain ? L'ouvrage Traitement automatique des langues naturelles. De l'analyse à l'application, rassemble des contributions issues de plusieurs champs disciplinaires comme la linguistique, l'informatique, la philosophie et l'analyse du discours. Il expose deux aspects primordiaux de tout traitement automatique : l'analyse et l'application informatique. Les analyses présentées sont de nature philosophique ou linguistique. Certaines de ces analyses s'inscrivent dans le cadre de la théorie des actes du discours, d'autres dans le cadre de théories linguistiques formelles. Les applications informatiques, quant à elles, bien que basées sur des analyses pour la plupart linguistiques, démontrent que cet aspect du traitement automatique des langues naturelles est primordial pour tester la robustesse des modèles et le bien-fondé des analyses.

International audience; Bibliometrics is more and more interested in the full text processing and the study of the structure of scientific papers. The contexts of in-text references present in articles are particularly relevant for such studies. This work describes the construction of the InTeReC dataset, which is an in-text reference corpus that aims to promote experimental reproducibility and to provide a standard dataset for further research. The InTeReC dataset is a set of sentences containing in-text references together with all the data necessary for their recontextualization in papers using standard CSV format. This should encourage the implementation of natural language processing tools for Bibliometric studies and related research in information retrieval and visualization.

The paper presents the results from the research study of students at the Faculty of Humanities and Social Sciences, University of Zagreb, Croatia about their awareness and exposure to well-known internet security threats as well as their awareness about security threats countermeasures. The results of the research study indicate high level of awareness of students about internet security threats as well as the fact that they are exposed to widely internet security known threats to a certain degree. While students demonstrated knowledge about application of security threats countermeasures, that part of their activities could be improved further. Generally, students need additional and updated knowledge to raise the level of their readiness in order to be able to respond to known and emerging internet security threats. The results of the research will be applied in planning of university courses related to the internet security.

International audience

This thesis focuses on the development of a tool for the automatic processing of Modern Standard Arabic, at the morphological and semanticlevels, with the final objective of Information Extraction on technological innovations. As far as the morphological analysis is concerned, our tool includes several successive processing stages that allow to label and disambiguate occurrences in texts : a morphological layer (Gibran 1.0), which relies on Arabic pattern as distinctive features ; a contextual layer (Gibran 2.0), which uses contextual rules ; and a third layer (Gibran 3.0), which uses a machine learning model. Our methodology is evaluated using the annotated corpus Arabic-PADT UD treebank. The evaluations obtain an F-measure of 0.92 and 0.90 for the morphological analyses. These experiments demontrate the possibility of improving such a corpus through linguistic analyses. This approach allowed us to develop a prototype of information extraction on technological innovations for the Arabic language. It is based on the morphological analysis and syntaxico-semantic patterns. This thesis is part of a doctoral-entrepreneurial course., Cette recherche porte sur le développement d’un outil de traitement automatique de la langue arabe standard moderne, au niveau morphologique et sémantique, avec comme objectif final l’extraction d’information dans le domaine de l’innovation technologique en entreprise. En ce qui concerne l’analyse morphologique, notre outil comprend plusieurs traitements successifs qui permettent d’étiqueter et de désambiguïser les occurrences dans les textes : une couche morphologique (Gibran 1.0), qui s’appuie sur les schèmes arabes commetraits distinctifs ; une couche contextuelle (Gibran 2.0), qui fait appel à des règles contextuelles ; et une troisième couche (Gibran 3.0) qui fait appel à un modèle d’apprentissage automatique. Notre méthodologie est évaluée sur le corpus annoté Arabic-PADT UD treebank. Les évaluations obtiennent une F-mesure de 0,92 et 0,90 pour les analyses morphologiques. Ces expérimentations montrent, entre autres, la possibilité d’améliorer une telle ressource par les analyses linguistiques. Cette approche nous a permis de développer un prototype d’extrac-tion d’information autour de l’innovation technologique pour la langue arabe. Il s’appuie sur l’analyse morphologique et des patrons syntaxico-sémantiques. Cette thèse s’inscrit dans un parcours docteur-entrepreneur.