logo EDITE Jerome DANTAN
Identité
Jerome DANTAN
État académique
Thèse en cours...
Sujet: Une approche systémique unifiée pour l'optimisation durable des système socio-environnementaux
Direction de thèse:
Laboratoire:
Voisinage
Ellipse bleue: doctorant, ellipse jaune: docteur, rectangle vert: permanent, rectangle jaune: HDR. Trait vert: encadrant de thèse, trait bleu: directeur de thèse, pointillé: jury d'évaluation à mi-parcours ou jury de thèse.
Productions scientifiques
oai:hal.archives-ouvertes.fr:hal-00869709
SEMANTIC INDEXATION OF WEB SERVICES FOR COLLABORATIVE EXPERT ACTIVITIES
In many research domains, scientists have to cope with large amounts of data issued from multiple sources, looking for regular patterns in raw data that may suggest models. The extraction of knowledge from data in general is widely used in fields such as biology, economics, etc. For this, they develop models in their respective area of research. The efficiency of the experts depend on their ability to cooperate: they share data and models. Indeed, each domain is both consumer of treatments provided by another team, and provider for the benefit of other areas. Each team of experts have to make available and publish business processes for the benefit of other teams, that allows them to have the most up to date treatments. In this work, we propose a work environment based on semantic web services indexation whose benefits will offer (1) a platform to publish and reuse treatments, accessible through the Internet, (2) services composition to chain treatments and assisted and/or automated selection of useful services. For this, we define two kinds of services: knowledge services, that provide business processes and data services that are wrappers that provide the relevant data from heterogeneous data sources. They will be accompanied with an existing OWL ontology, used by the experts of the considered domain. As a result, experts will be able to operate and cooperate through heterogeneous application environments and all around the world. An example on modeling farm durability indicators will be presented.
PROCEEDINGS OF THE IADIS INTERNATIONAL CONFERENCE INFORMATION SYSTEMS 2012 IADIS INTERNATIONAL CONFERENCE INFORMATION SYSTEMS 2012conference proceeding 2012-03-10
oai:hal.archives-ouvertes.fr:hal-00869722
A KDD PROCESS TO RETRIEVE AND AGGREGATE DATA FROM RELATIONAL DATABASES
Relational databases are a standard for representing data models. SQL is the most widely used language for querying such databases. Consequently, in many research domains, scientists extract data from relational databases, compute them and do statistical treatments. But they have to deal with the complexity of relational databases models. In addition, it takes a long time for the scientists to manually retrieve and compute data. That's why we propose a system which automatically does. It contains the following layers: parameterizable extraction of data, automatic process of SQL queries, data aggregation, statistical parameters computation, writing the results to tables and final data processing by the scientist, thanks to a statistical analysis software. A use case on the research on a soil quality index from a large relational database will be presented.
PROCEEDINGS OF THE IADIS INTERNATIONAL CONFERENCE INFORMATION SYSTEMS 2012 IADIS INTERNATIONAL CONFERENCE INFORMATION SYSTEMS 2012conference proceeding 2012-03-10
oai:hal.archives-ouvertes.fr:hal-00869735
The G.O.A.L. Approach
In a global context of sharing information, such as Big Data and cloud computing paradigms, researchers are developing many means to deal with new data models and algorithms. However, the development and reuse of these ones is complex because of the heterogeneity of environments, data formats and contexts of use all around of the world. That's why a way to share and reuse algorithms and treatments through a common formalism is needed, for both machines and computers. The ultimate goal of our work is to provide a collaborative platform for not only experts but also machines automatically develop, reuse and chain treatments, computations and models. For this, we rely on a goal-oriented approach which is associated with the Semantic Web, to establish a common formalism to design models for worldwide researchers. In this article, we propose the formalization of our approach thanks to an algebra which is linked to the Semantic Web standards. Finally, we provide a high-level language dedicated to both computers and experts, illustrated with examples that are linked to the agriculture domain.
ENASE 2013 - Proceedings of the 8th International Conference on Evaluation of Novel Approaches to Software Engineering ENASE 2013 - 8th International Conference on Evaluation of Novel Approaches to Software Engineeringconference proceeding 2013-07-04
oai:hal.archives-ouvertes.fr:hal-01018063
Taking Account of Uncertain, Imprecise and Incomplete Data in Sustainability Assessments in Agriculture
To manipulate incomplete, uncertain and hypothetic information about farms, the future evaluation software in the agriculture domain will have to make computations and combine fuzzy numerical values. The approach we propose relies on combination of incomplete, uncertain and hypothetic object values based on particular application of the Dempster-Shafer evidence theory. We then propose the integration of this approach in classic object oriented programming via the implementation of a C++ library.
Lecture notes in computer science Lecture Notes in Computer Science LNCS Computational Science and Its Applications - ICCSA 2014 - 14th International Conferenceconference proceeding 2014-06-30
oai:hal.archives-ouvertes.fr:hal-01018088
A goal-oriented meta-model for scientific research
In many research domains, studies need to be addressed in a multidisciplinary manner: each expert deals with a particular aspect of the problem. It may be useful for experts to take into account new data, share partial results, and update their own indicators and models, in order to take advantage of new measures and updated indexes in real time. For this, an easily-understandable knowledge model for any raw data source, statistical operator, indicator or business process to be available for experts, is needed. In this paper, we propose a goal-oriented meta-model, to index and reuse treatments and an extension of existing semantic Web standards to index goal-oriented services and assist and/or automate their selection. These features enable capabilities for interoperability and information exchange between three layers of knowledge: goal, domain and data layers. An application with an existing ontology of the agriculture domain and farm durability indicators will be proposed.
Lecture notes in computer science Lecture Notes in Computer Science LNCS 8583 Computational Science and Its Applications - ICCSA 2014 - 14th International Conferenceconference proceeding 2014-06-30
oai:hal.archives-ouvertes.fr:hal-01214367
Combination of Imperfect Data in Fuzzy and Probabilistic Extension Classes
International audience
In this article, we propose a uniform formal model able to handle uncer-tain data. The approach presented provides a formalism for both representing and manipulating rigorously quantities which may have a finite number of possible or probable values with their interdependencies. Then, we define an algebraic structure to operate chained computations on such quantities with properties similar to R. Next, we provide a par-ticular interpretation for mixing such quantities through the Dempster-Shafer theory. Finally, we provide an implementation of this approach into object oriented programming.
ISSN: 2325-6192 Journal of Environmental Accounting and Management https://hal.archives-ouvertes.fr/hal-01214367 Journal of Environmental Accounting and Management, L & H Scientific Publishing, 2015, 3 (2), <https://lhscientificpublishing.com/Journals/JEAM-Default.aspx>. <10.5890/JEAM.2015.06.004> https://lhscientificpublishing.com/Journals/JEAM-Default.aspxARRAY(0x7f5471618770) 2015-06
oai:hal.archives-ouvertes.fr:hal-01380991
A formal model to compute uncertain continuous data
International audience
Current researches in the domain of Information and Communication Technologies describe and extend the existing formalisms to develop systems that compute uncertain data. Indeed, handling uncertain data is a great challenge for complex systems. In this article, we provide a formal model to compute such data rigorously. Such quantities may be interpreted as either possible or probable values, added to their interdependencies. For this, the algebraic structure we defined is a vector space. We then provide a particular way for mixing such continuous quantities.
In proceedings of CS-DC'15 World e-conference (Complex Systems Digital Campus) UNITWIN/UNESCO CS-DC'15 World e-conference https://hal.archives-ouvertes.fr/hal-01380991 CS-DC'15 World e-conference, Sep 2015, e-conference, France. In proceedings of CS-DC'15 World e-conference (Complex Systems Digital Campus) UNITWIN/UNESCO, 2015, <http://cs-dc-15.org/> http://cs-dc-15.org/ARRAY(0x7f547246fe30) 2015-09-30
oai:hal.archives-ouvertes.fr:hal-01380999
A formal approach for the representation and the combination of imperfect data.
International audience
Nowadays, the sustainability of human activities is a major worldwide concern. Indeed, the problem is no longer to evaluate only the efficiency of human activities, but also sustainability along many axes that can be of various kinds: economic, social, environmental, etc. Such assessments are a major challenge for today’s society. Because of the exponential development of means of data recording and storage (“big data” buzz word), and on the basis of Volume, Variety and Velocity properties of big data, scientists need to compute large amounts of data and so do not necessarily have time to clean them. In this context, they compute all available data whose types of imperfections are heterogeneous. Actors in several domains have to cope with such data, especially to assist humans in their decisions by merging them from many data sources (e.g. measurements, sensors, observations) to model behaviours of complex systems. Mathematical approaches to model imperfect data are well known and established in various scientific areas today, such as both probability based and possibility based calculus. Decisions of experts from various fields have to handle rigorous computations and aggregations of both data and their associated uncertainty. We propose a rigorous model to handle uncertainty on the attributes of objects, and a way to rigorously aggregate discrete data, whose imperfections nature are covered either by the classical probability theory (randomness), either by the possibility theory (fuzziness) thanks to the Dempster-Shafer theory.
Agrostat 2016 congress, March 21-24 2016, Lausanne, Switzerland Agrostat 2016 https://hal.archives-ouvertes.fr/hal-01380999 Agrostat 2016, Mar 2016, Lausanne, Switzerland. Agrostat 2016 congress, March 21-24 2016, Lausanne, Switzerland, <http://agrostat2016.sfds.asso.fr/e-proceedings/> http://agrostat2016.sfds.asso.fr/e-proceedings/ARRAY(0x7f5472471578) 2016-03-21
oai:hal.archives-ouvertes.fr:hal-01380996
A systemic meta-model for socio-environmental systems
International audience
We propose a systemic meta-model for the sustainable simulation of socio-environmental complex systems. The approach presented integrates data uncertainty management, for both representing and manipulating rigorously quantities which may have a finite number of possible or probable values with their interdependencies. We also provide an operationalization of such models for both data retrieving, via an object-relational mapping, and model simulation, via series of triples, which are linked to examples in the field of agriculture.
CSD&M 2015 (International Conference on Complex Systems Design & Management) https://hal.archives-ouvertes.fr/hal-01380996 Auvray, G., Bocquet, J.-C., Bonjour, E., Krob, D. CSD&M 2015 (International Conference on Complex Systems Design & Management), Nov 2015, Paris, France. In proceedings of the Sixth International Conference on Complex Systems Design & Management, CSD&M 2015., <http://www.csdm2015.csdm.fr/> http://www.csdm2015.csdm.fr/ARRAY(0x7f547246cad0) 2015-11-23