Éditeur(s) : HAL CCSD
Résumé : International audience
It is now well recognized that information management systems often deal with geographic information even if they often do not take advantage of it. In these systems, requirements engineering (RE) is becoming more and more important in helping stakeholders to model their requirements regarding the quality of the data.However, to date, few RE projects take geographic information and geographic constraints into account. We thus propose to study in this paper the question of managing such geographic constraints in the framework of RE. We propose an extension of the RE model and its implementation within the KAOS framework, and the use of geographic ontologies
2nd International Conference on Knowledge Management, Information and Knowledge Systems
Hammamet, Tunisia

lirmm-01230140
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01230140

Éditeur(s) : HAL CCSD
Résumé : International audience
After defining the High Performance Pre- Computing --referred as HPPC-- concept, the aim of the present study is to develop a prototype whether to approve or not the benefits of this concept. Our application case tries to answer the geophysical issue of coastal flooding. This is an example of an alert system based on the HPPC architecture, thus on pre-computed scenarios. The prototype provides the scientists with an ergonomic and on-demand tool allowing the run of scenarios of any implemented numerical models. These runs are available through a web application which submits the corresponding jobs on the remote french public cluster of HPC@LR. In this study we simulate the waves propagation over a Mediterranean grid using the wave model WaveWatch III⃝R . A reference simulation using usual conditions is approximated using the k-NN algorithm over 12, 98 and then 980 pre-computed scenarios. This simple experiment demonstrates how useful the pre-computing of scenarios is for alert systems as far as enough and relevant scenarios are pre-computed. This is the reason why searches continue in each critical points of the HPPC architecture such as the design of experiment, the approximation of the results by meta-models and the research of the closest scenarios in this big data context.
4th International Conference on Knowledge and Systems Engineering
Danang, Vietnam

lirmm-00736797
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00736797
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00736797/document
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00736797/file/PID2396685.pdf
DOI : 10.1109/KSE.2012.36

Éditeur(s) : HAL CCSD SciTePress
Résumé : International audience
Data integration and knowledge capitalization combine data and information coming from different data sources designed by different experts having different purposes. In this paper, we propose to assist the underlying model merging activity. For close models made by experts of various specialities, we partially automate the identification of a Greatest Common Model (GCM) which is composed of the common concepts (core concepts) of the different models. Our methodology is based on Formal Concept Analysis which is a method of data analysis based on lattice theory. A decision tree allows to semi-automatically classify concepts from the concept lattices and assist the GCM extraction. We apply our approach on the EIS-Pesticide project, an environmental information system which aims at centralizing knowledge and information produced by different specialized teams.
ICEIS 2012: 14th International Conference on Enterprise Information Systems
Wroclaw, Poland

lirmm-00727009
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00727009

Éditeur(s) : HAL CCSD
Résumé : International audience
Data integration and knowledge capitalization combine data and information coming from different data sources designed by different experts having different purposes. In this paper, we propose to assist the underlying model merging activity. For close models made by experts of various specialities on the same system, we partially automate the identification of a Greatest Common Model (GCM) which is composed of the common concepts (core-concepts) of the different models. Our methodology is based on Formal Concept Analysis which is a method of data analysis based on lattice theory. A decision tree allows to semiautomatically classify concepts from the concept lattices and assist the GCM extraction. We apply our approach on the EIS-Pesticide project, an environmental information system which aims at centralizing knowledge and information produced by different research teams.
14th International Conference on Enterprise Information Systems (ICEIS 2012)
Wroclaw, Poland

lirmm-01319858
http://hal-lirmm.ccsd.cnrs.fr/lirmm-01319858
DOI : 10.1007/978-3-642-40654-6_5

Éditeur(s) : HAL CCSD Cépaduès-Éditions
Résumé : International audience
Nous présentons dans cet article un nouvel algorithme permettant la construction et la mise à jour incrémentale du FIAθ : FIASCO. Notre algorithme effectue un seul passage sur les données et permet de prendre en compte les nouveaux batches, itemset par itemset et pour chaque itemset, item par item.
EGC: Extraction et Gestion des Connaissances
Nice, France

hal-00447421
https://hal.archives-ouvertes.fr/hal-00447421
https://hal.archives-ouvertes.fr/hal-00447421/document
https://hal.archives-ouvertes.fr/hal-00447421/file/2008--Lionel_Vinceslas--FIASCO_un_nouvel_algorithme_d_extraction_d_itemsets_frequents_dans_les_flots_de_donnees.pdf

Éditeur(s) : HAL CCSD
Résumé : International audience
Mining sequential patterns in data streams is a new challenging problem for the datamining community since data arrives sequentially in the form of continuous rapid and infinite streams. In this paper, we propose a new on-line algorithm, SPAMS, to deal with the sequential patterns mining problem in data streams. This algorithm uses an automaton-based structure to maintain the set of frequent sequential patterns, i.e. SPA (Sequential Pat- tern Automaton). The sequential pattern automaton can be smaller than the set of frequent sequential patterns by two or more orders of magnitude, which allows us to overcome the problem of combinatorial explosion of se- quential patterns. Current results can be output constantly on any user 's specified thresholds. In addition, taking into account the characteristics of data streams, we propose a well-suited method said to be approximate since we can provide near optimal results with a high probability. Experimental studies show the relevance of the SPA data structure and the efficiency of the SPAMS algorithm on various datasets. Our contribution opens a promis- ing gateway, by using an automaton as a data structure for mining frequent sequential patterns in data streams.
Advances in Knowledge Discovery and Management

lirmm-00435841
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00435841
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00435841/document
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00435841/file/akdmSPAMS.pdf

Éditeur(s) : HAL CCSD
Résumé : LPSC14037
Scientific data collected with modern sensors or dedicated detectors exceed very often the perimeter of the initial scientific design. These data are obtained more and more frequently with large material and human efforts. A large class of scientific experiments are in fact unique because of their large scale, with very small chances to be repeated and to superseded by new experiments in the same domain: for instance high energy physics and astrophysics experiments involve multi-annual developments and a simple duplication of efforts in order to reproduce old data is simply not affordable. Other scientific experiments are in fact unique by nature: earth science, medical sciences etc. since the collected data is "time-stamped" and thereby non-reproducible by new experiments or observations. In addition, scientific data collection increased dramatically in the recent years, participating to the so-called "data deluge" and inviting for common reflection in the context of "big data" investigations. The new knowledge obtained using these data should be preserved long term such that the access and the re-use are made possible and lead to an enhancement of the initial investment. Data observatories, based on open access policies and coupled with multi-disciplinary techniques for indexing and mining may lead to truly new paradigms in science. It is therefore of outmost importance to pursue a coherent and vigorous approach to preserve the scientific data at long term. The preservation remains nevertheless a challenge due to the complexity of the data structure, the fragility of the custom-made software environments as well as the lack of rigorous approaches in workflows and algorithms. To address this challenge, the PREDON project has been initiated in France in 2012 within the MASTODONS program: a Big Data scientific challenge, initiated and supported by the Interdisciplinary Mission of the National Centre for Scientific Research (CNRS). PREDON is a study group formed by researchers from different disciplines and institutes. Several meetings and workshops lead to a rich exchange in ideas, paradigms and methods. The present document includes contributions of the participants to the PREDON Study Group, as well as invited papers, related to the scientific case, methodology and technology. This document should be read as a "facts finding" resource pointing to a concrete and significant scientific interest for long term research data preservation, as well as to cutting edge methods and technologies to achieve this goal. A sustained, coherent and long term action in the area of scientific data preservation would be highly beneficial.
http://hal.in2p3.fr/in2p3-00959072

in2p3-00959072
http://hal.in2p3.fr/in2p3-00959072
http://hal.in2p3.fr/in2p3-00959072/document
http://hal.in2p3.fr/in2p3-00959072/file/PREDON_Scientific_Data_Preservation_2014.pdf

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Addressing data quality issues in information systems remains a challenging task. Many approaches only tackle this issue at the extract, transform and load steps. Here we define a comprehensive method to gain greater insight into data quality characteristics within data warehouse. Our novel architecture was implemented for an hydroecological case study where massive French watercourse sampling data are collected. The method models and makes effective use of spatial, thematic and temporal accuracy, consistency and completeness for multidimensional data in order to offer analysts a “data quality” oriented framework. The results obtained in experiments carried out on the Saône River dataset demonstrated the relevance of our approach
ISSN: 0098-3004

hal-01223918
https://hal.archives-ouvertes.fr/hal-01223918
https://hal.archives-ouvertes.fr/hal-01223918/document
https://hal.archives-ouvertes.fr/hal-01223918/file/final-cogeo.pdf
DOI : 10.1016/j.cageo.2015.09.012

Éditeur(s) : HAL CCSD International Frequency Sensor Association (IFSA)
Résumé : International audience
This paper introduces an approach illustrating how the Open Geospatial Consortium (OGC) Sensor Web Enablement (SWE) framework can be used in order to build a franco-lebanese observatory. We present the practical application of SWE services as a source of real-time observation data and the associated technical architecture for making near real-time observations available to end users on the Web. We discuss the question of crossing sensor data with other data sources, e.g., data provided by human observations. We illustrate our approach by describing the methodology to integrate a first illustration case to monitor snow weather stations in the Lebanese Mountains.
ISSN: 1726-5479

lirmm-01381092
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01381092

Éditeur(s) : HAL CCSD
Résumé : Séminaire de Synthèse du Projet FPPSM "Forêts, Parcs, Pauvreté au Sud de Madagascar"
A Madagascar, la déforestation mobilise toutes les recherches, mais il manque encore une base de données complète, référencée et réactualisable sur les dynamiques forestières. Une solution est la valorisation d'images satellitaires multidates qui, de ce fait, sont issues de capteurs différents en termes de résolution spatiale, radiométrique et temporelle (Spot, Landsat, Modis). Ces données géolocalisées sont volumineuses, elles doivent être normalisées puis stockées et finalement traitées. Face à l'hétérogénéité des données et traitements, l'interopérabilité des ressources est assurée par le standard ISO (International Standard Organisation) et la spécification OGC (Open Geospatial Consortium). Le traitement de séries chronologiques de la couverture forestière exige un outil réutilisable qui s'adapte à de nouvelles données satellitaires : des chaînes de traitement pour les données simples, des workflows pour les gros volumes de données. La modélisation du workflow est illustrée par la construction de sa base de connaissance en utilisant un contexte de travail.
https://hal.archives-ouvertes.fr/hal-01370017

hal-01370017
https://hal.archives-ouvertes.fr/hal-01370017
IRD : fdi:010066424

Éditeur(s) : HAL CCSD
Résumé : National audience
Lorsque le volume des données est trop important pour qu'elles soient stockées dans une base de données, ou lorsque leur taux d'arrivée est rapide, les Systèmes de Gestion de Flux de Données (SGFD) permettent de capturer des flux d'enregistrements structurés et de les interroger à la volée par des requêtes permanentes (exécutées de façon continue). Mais les SGFD ne conservent pas l'historique des flux qui est perdu à jamais. Cette communica- tion propose une définition formelle de ce que devrait être un résumé généraliste de flux de données. La notion de résumé généraliste est liée à la capacité de répondre à des requêtes variées et de réaliser des tâches variées de fouille de données, en utilisant le résumé à la place du flux d'origine. Une revue de plusieurs approches de résumés est ensuite réalisée dans le cadre de cette définition. Enfin, l'intégration de telles approches de résumés dans un SGFD est discutée, à la fois en termes de construction et d'utilisation.
EGC: Extraction et Gestion des Connaissances
Hammamet, Tunisia

hal-00502017
https://hal.archives-ouvertes.fr/hal-00502017
https://hal.archives-ouvertes.fr/hal-00502017/document
https://hal.archives-ouvertes.fr/hal-00502017/file/resume_generaliste.pdf

Éditeur(s) : HAL CCSD ACM
Résumé : International audience
Environmental resources (e.g., air quality, water quantity) are needed to understand fundamental questions such as global change. Such resources are often collected from sensors, including humans acting as sensors. Tools have emerged to manage such data in the form of time series and, in particular, the Sensor Observation Service (SOS) which offers a framework based on predefined relational database schema. Environmental observatories can be built using such frameworks, allowing to address specific key scientific questions by collecting and sharing large-scale environmental data. However, the strict schema of SOS database makes it difficult to integrate some data that cannot be directly mapped to the schema. Guidelines and best practices are offered in the literature in order to reuse standards from the Semantic Web but they do not cover all needs. In particular, they do not help to reflect the fact that a single environmental database can lead to several SOS models. Since being aware of these multiple possibilities is crucial for a better use of the observatories, we argue that some extensions of the existing works are required. In this paper, we thus propose an extension of existing vocabularies to achieve this goal. Our contribution is illustrated on the real case of the Lebanese-French O-LiFE environmental observatory.
6th ACM International Conference on Web Intelligence, Mining and Semantics
Nimes, France

lirmm-01381082
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01381082
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01381082/document
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01381082/file/presentationwims_s1_2.pdf
DOI : 10.1145/2912845.2912856

Éditeur(s) : HAL CCSD
Résumé : International audience
In this paper, we investigate image segmentation by region merging. Given any similarity measure between regions , satisfying some weak constraints, we give a general predicate for answering if two regions are to be merged or not during the segmentation process. Our predicate is generic and has six properties. The first one is its inde-pendance with respect to the similarity measure, that leads to a user-independant and adaptative predicate. Second, it is non-parametric, and does not rely on any assumption concerning the image. Third, due to its weak constraints, knowledge may be included in the predicate to fit better to the user's behaviour. Fourth, provided the similarity is well-chosen by the user, we are able to upperbound one type of error made during the image segmentation. Fifth, it does not rely on a particular segmentation algorithm and can be used with almost all region-merging algorithms in various application domains. Sixth, it is calculated quickly, and can lead with appropriated algorithms to very efficient segmentation.
10th IEEE International Conference on Tools with Artificial Intelligence
Taipei, Taiwan

lirmm-01168371
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01168371
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01168371/document
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01168371/file/FioNocICTAI98_Image%20Segmentation%20Using.pdf
DOI : 10.1109/TAI.1998.744885

Éditeur(s) : HAL CCSD BMJ Publishing Group
Résumé : International audience
OBJECTIVE:To identify local meteorological drivers of dengue fever in French Guiana, we applied an original data mining method to the available epidemiological and climatic data. Through this work, we also assessed the contribution of the data mining method to the understanding of factors associated with the dissemination of infectious diseases and their spatiotemporal spread.METHODS:We applied contextual sequential pattern extraction techniques to epidemiological and meteorological data to identify the most significant climatic factors for dengue fever, and we investigated the relevance of the extracted patterns for the early warning of dengue outbreaks in French Guiana.RESULTS:The maximum temperature, minimum relative humidity, global brilliance, and cumulative rainfall were identified as determinants of dengue outbreaks, and the precise intervals of their values and variations were quantified according to the epidemiologic context. The strongest significant correlations were observed between dengue incidence and meteorological drivers after a 4-6-week lag.DISCUSSION:We demonstrated the use of contextual sequential patterns to better understand the determinants of the spatiotemporal spread of dengue fever in French Guiana. Future work should integrate additional variables and explore the notion of neighborhood for extracting sequential patterns.CONCLUSIONS:Dengue fever remains a major public health issue in French Guiana. The development of new methods to identify such specific characteristics becomes crucial in order to better understand and control spatiotemporal transmission.Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.
ISSN: 1067-5027

lirmm-01379072
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01379072
DOI : 10.1136/amiajnl-2013-002348
PUBMED : 24549761

Éditeur(s) : HAL CCSD Cépaduès-Éditions
Résumé : International audience
Nous présentons dans cet article un nouvel automate : le FIA qui permet de traiter de façon efficace la problématique de l'extraction des itemsets fréquents dans les data streams. Le FIA est une structure de données très compacte et informative qui, par ailleurs, présente des propriétés incrémentales facilitant grandement la mise à jour avec une granularité très fine. L'algorithme développé pour la mise à jour du FIA effectue un unique passage sur les données qui sont prises en compte par batch, itemset par itemset et pour chaque itemset, item par item. Utilisé dans le cadre d'une approche prédictive, le FIA permet d'indexer les itemsets véritablement fréquents du stream en maximisant soit le rappel soit la précision.
EGC: Extraction et Gestion des Connaissances
Nice, France

hal-00447420
https://hal.archives-ouvertes.fr/hal-00447420
https://hal.archives-ouvertes.fr/hal-00447420/document
https://hal.archives-ouvertes.fr/hal-00447420/file/2008--Jean-Emile_Symphor--Le_FIA_un_nouvel_automate_permettant_l_extraction_efficace_d_itemsets_frequents_dans_les_data_streams.pdf

Éditeur(s) : HAL CCSD
Résumé : International audience
9th International Conference on Sensor Technologies and Applications
Venise, Italy

lirmm-01381091
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01381091

Éditeur(s) : HAL CCSD
Résumé : International audience
In the analysis of coastal hazards, the features of extreme waves are determining information to question the impact of storms to the coast. The spatial behaviour of extreme waves is even more valuable especially since it is sparsely provided. Regarding recent applications in other contexts, a kind of statistical models called max-stable processes is relevant for modelling spatial extreme events. Max-stable processes are extensions of the well-known Generalised Extreme Value (GEV) distribution. Unlike univariate approaches, max-stable processes consider spatial dependence of a phenomenon. Such a modelling also overtakes a standard multivariate approach by providing information continuously over the area studied, even where no observation is available. Relying on such a stochastic modelling, the aim of this study is to discuss the extreme waves hazards in the Gulf of Lions, focusing on their spatial behaviour.
34th Conference on Coastal Engineering
Seoul, South Korea

lirmm-01382568
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01382568
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01382568/document
https://hal-lirmm.ccsd.cnrs.fr/lirmm-01382568/file/7553-31212-1-PB.pdf
DOI : 10.9753/icce.v34.management.17

Éditeur(s) : HAL CCSD
Résumé : International audience
L'extraction de motifs séquentiels fréquents dans les data streams est un enjeu important traité par la communauté des chercheurs en fouille de données. Plus encore que pour les bases de données, de nombreuses contraintes supplémentaires sont à considérer de par la nature intrinsèque des streams. Dans cet article, nous proposons un nouvel algorithme en une passe : SPAMS, basé sur la construction incrémentale, avec une granularité très fine par transaction, d'un automate appelé SPA, permettant l'extraction des motifs séquentiels dans les streams. L'information du stream est apprise à la volée, au fur et à mesure de l'insertion de nouvelles transactions, sans pré-traitement a priori. Les résultats expérimentaux obtenus montrent la pertinence de la structure utilisée ainsi que l'efficience de notre algorithme appliqué à différents jeux de données.
EGC'09 : Extraction et Gestion des Connaissances
Strasbourg, France

lirmm-00345563
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00345563

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
A data stream is a potentially uninterrupted flow of data. Mining this flow makes it necessary to cope with uncertainty, as only a part of the stream can be stored. In this paper, we evaluate a statistical technique which biases the estimation of the support of patterns, so as to maximize either the precision or the recall, as chosen by the user, and limit the degradation of the other criterion. Theoretical results show that the technique is not far from the optimum, from the statistical standpoint. Experiments performed tend to demonstrate its potential, as it remains robust even under significant distribution drifts.
ISSN: 0031-3203

lirmm-00365474
https://hal-lirmm.ccsd.cnrs.fr/lirmm-00365474

Éditeur(s) : HAL CCSD
Résumé : National audience
Les entrepôts de données occupent aujourd'hui une place centrale dans le processus décisionnel. Outre leur consultation, une des finalités des entrepôts est de servir de socle aux techniques de fouilles de données. Malheureusement, les approches existantes exploitent peu les particularités des entrepôts (multidimensionnalité, hiérarchies et données historiques). Parmi ces méthodes, l'extraction de motifs séquentiels multidimensionnels a récemment été étudiée. Nous montrons dans cet article que ces dernières ne tirent pas pleinement profit des hiérarchies et ne découvrent par conséquent qu'une partie seulement des motifs qualitativement intéressants. Nous proposons alors une méthode d'extraction de motifs séquentiels multidimensionnels basée sur un automate et extrayant de nouveaux motifs. Les différentes expérimentations menées sur des jeux de données synthétiques attestent des bonnes performances de notre proposition.
EGC: Extraction et Gestion des Connaissances
Hammamet, Tunisia

hal-00502002
https://hal.archives-ouvertes.fr/hal-00502002
https://hal.archives-ouvertes.fr/hal-00502002/document
https://hal.archives-ouvertes.fr/hal-00502002/file/SALINES-EGC.pdf