Search

The Saône and the Rhône rivers have experienced and are still experiencing major floods which complicate the development of the neighbouring territories. The Plan Rhône managed by the Direction Régionale de l’Environnement, de l’Aménagement et du Logement Auvergne Rhône-Alpes (DREAL ARA), is an interregional contract whose one objective is to reduce vulnerability to flood risk. The construction of protection structures - also known as dikes - is the main way to protect against floods. Currently, the dispersion and heterogeneity of information relating to these structures between many stakeholders complicates the task of the DREAL's Pôle Plan Rhône, of which one of the objectives is to have an overall view of all these structures.This is why the DREAL ARA wanted to create a centralized knowledge base of the geographical data available on the dikes and include them into a computer tool that would allow to visualize the data and operate spatial analysis. The first results of an inventory and a preliminary critical analysis of the available GIS data, carried out by SEPIA Conseils, were essential for the data gathering, which is the real subject of this document.How, with the help of the results of the previous phases, to gather all the information from the available source databases by integrating both geometric objects and attribute information in an exhaustive way and without losing any information? How to ensure the reliability of this knowledge base in the short but also in the medium term? How to design this specific tool that would make it easy for users to use?The development of this database required the set up of specific methodologies combining semi-automatic processing and manual verification in order to deal with specific cases. Validation and control steps were implemented during the conception of the base, and during its finalization in order to make the base more reliable.The tool for visualizing and editing this knowledge base relies on a visualization interface in the free software QGIS (Long Term Release 3.4 version) and a PostgreSQL/PostGIS database. This tool allows the visualization, editing, modification and the production of customizable synthesis maps. It was created according to the needs identified, in order to be adapted and to ensure the durability of its use over time.In addition, a prioritized investigation plan has been drawn up in order to make the dike position more reliable. A financial report of these additional investigations and summary sheets of the selected sections allowed to study the financial and material feasibility in the short and medium term.Because of lack of information three of the eight tables could not be completed during this study. They should be completed in the future, by the DREAL ARA itself or by other users. This additional research will be essential to ensure the sustainability of the DREAL knowledge base and the associated tool.; Le Rhône et la Saône sont des fleuves qui ont connu et connaissent toujours des crues importantes, qui compliquent le développement des territoires qu’ils traversent. Le Plan Rhône, piloté par la DREAL Auvergne Rhône-Alpes, est un contrat interrégional dont un des objectifs est de réduire la vulnérabilité des territoires face au risque d’inondation. Un des moyens principaux pour se protéger contre les crues est la construction d’ouvrages de protection, appelés aussi plus communément digues. La dispersion et l’hétérogénéité des informations relatives à ces ouvrages entre de nombreux acteurs complexifie à l’heure actuelle la tâche du Pôle Plan Rhône de la DREAL, dont un des objectifs est d’avoir une vision globale de l’ensemble de ces ouvrages. C’est la raison pour laquelle la DREAL a souhaité la constitution d’une base de connaissances centralisée des données géographiques disponibles sur les digues et leur intégration dans un outil informatique permettant leur visualisation et la production d’analyses spatiales. Les premiers résultats d’un recensement et d’une analyse critique préliminaire des données SIG disponibles réalisés par SEPIA Conseils ont été indispensables à la centralisation des données, réel objet de ce mémoire.Comment, avec l’aide des résultats des phases précédentes, rassembler l’ensemble des informations des bases sources récupérées en intégrant à la fois les objets géométriques mais aussi les informations attributaires de manière exhaustive et sans perdre d’information ? Comment fiabiliser une telle base de connaissance à court mais aussi à moyen terme ? Comment concevoir cet outil spécifique qui permettrait de la rendre facilement exploitable pour les utilisateurs ?L’élaboration de cette base a nécessité de développer des méthodologies spécifiques couplant traitements semi-automatiques et vérification manuelles afin de traiter les cas particuliers. Des étapes de validation et de contrôle ont été mises en place au cours de la construction de la base, et lors de sa finalisation afin de fiabiliser la base.L’outil de visualisation et d’édition de cette base de données, s’appuie sur une architecture couplant une interface de visualisation dans le logiciel libre QGIS (version Long Term Release 3.4) et une base de données PostgreSQL/PostGIS. Cet outil permet la visualisation, la consultation, l’édition, la modification et la production de cartographies de synthèse personnalisables. Il a été créé selon les besoins identifiés afin d’être bien adapté et d’assurer la durabilité d’utilisation dans le temps.Un plan d’investigations priorisé a été élaboré en complément afin de fiabiliser le tracé et la connaissance de certains tronçons de digue. Un chiffrage de ces investigations complémentaires et des fiches synthèses des tronçons sélectionnés ont permis d’étudier la faisabilité financière et matérielle d’une telle fiabilisation à court et moyen terme. Les différentes tables qui n’ont pas pu être complétées lors de cette étude par manque d’information devront l’être a posteriori, par la DREAL elle-même ou d’autres utilisateurs. Ces recherches complémentaires seront indispensables pour pérenniser la base de connaissances et l’outil associé.

Session 3: Systems Biology Session 3: Systems Biology; International audience

International audience; Ontologies are a well-motivated formal representation to model knowledge needed to extract and encode data from text. Yet, their tight integration with Information Extraction (IE) systems is still a research issue, a fortiori with complex ones that go beyond hierarchies. In this paper we introduce an original architecture where IE is specified by designing an ontology, and the extraction process is seen as an Ontology Population (OP) task. Concepts and relations of the ontology define a normalized text representation. As their abstraction level is irrelevant for text extraction, we introduced a Lexical Layer (LL) along with the ontology, i.e. relations and classes at an intermediate level of normalization between raw text and concepts. On the contrary to previous IE systems, the extraction process only involves normalizing the outputs of Natural Language Processing (NLP) modules with instances of the ontology and the LL. All the remaining reasoning is left to a query module, which uses the inference rules of the ontology to derive new instances by deduction. In this context, these inference rules subsume classical extraction rules or patterns by providing access to appropriate abstraction level and domain knowledge. To acquire those rules, we adopt an Ontology Learning (OL) perspective, and automatically acquire the inference rules with relational Machine Learning (ML). Our approach is validated on a genic interaction extraction task from a Bacillus subtilis bacterium text corpus. We reach a global recall of 89.3% and a precision of 89.6%, with high scores for the ten conceptual relations in the ontology

International audience; Information Extraction (IE) systems have been proposed in recent years, to extract genic interactions from bibliographical resources. But they are limited to single interaction relations, and have to face a tradeoff between recall and precision, by focusing either on specific interactions (for precision), or general and unspecified interactions of biological entities (for recall). Yet, biologists need to process more complex data from literature, in order to study biological pathways, so an ontology is an adequate formal representation to model this sophisticated knowledge. But the tight integration of IE systems and ontologies is still a current research issue, a fortiori with complex ones that go beyond hierarchies. Here, we propose a rich modeling of genic interactions with an ontology, and show how it can be used within an IE system. The ontology is seen as a language specifying a normalized representation of text. IE is performed by first extracting instances from Natural Language Processing (NLP) modules, then deductive inferences on the ontology language are completed. New instances may be infered, bringing together otherwise scattered textual information. We validated our approach on an annotated corpus of gene transcription regulations in Bacillus subtilis. We reach a global recall of 89.3% and a precision of 89.6%, with high scores for the ten semantic relations defined in the ontology

International audience; This paper presents an information system developed to help the assessment of the microbiological risk in food products. UQS (Unified Querying System) is composed of two distinct bases (a relational database and a conceptual graph knowledge base) which are integrated by means of a uniform querying language. The specificity of the system is that both bases include fuzzy data. Moreover, UQS allows the expression of preferences into the queries, by means of the fuzzy set theory.

Communication présentée au 11th International Conference on Conceptual Structures, ICCS 2003, Dresden, Germany, July 21-25, 2003; International audience; In the context of a microbiological application, our study proposes to extend the Conceptual Graph Model in order to allow one: (i) to represent imprecise data and queries that include preferences, by using fuzzy sets (from fuzzy set theory) in concept vertices, in order to describe either an imprecise concept type or an imprecise referent; (ii) to query a conceptual graph that may include imprecise data (factual graph) using a conceptual graph that may include preferences (query graph). This is performed in two steps: firstly by extending the projection operation to fuzzy concepts, secondly by defining a comparison operation characterised by two matching degrees: the possibility degree of matching and the necessity degree of matching between two graphs, and particularly between a query graph and a factual graph.

National audience