Difference between revisions of "François de Bertrand de Beuvron"
| Line 1: | Line 1: | ||
| − | == | + | == Research == |
| − | + | In the generic field of knowledge representation, my work aims to define an hybrid inference formalisms based on description logics and rules, and suitable for the formulation of problems related to the design, configuration, diagnosis, or classification of systems. | |
| − | ; | + | ; Semantic interpretation of high resolution satellite image. |
| − | + | To enable the identification of high-level structured objects (house, street ...), it is necessary to identify and merge segmented regions from image analysis. A geographic ontology describing not only the thematic objects but also their qualitative and quantitative spatial relationships is used to guide this interpretation. | |
| − | ; | + | ; Hybrid representation of knowledge for innovative design. |
| − | + | The conceptual phase of innovative design is discussed in the context of the TRIZ theory for which an ontology of problem-solving concepts is proposed. The aim is to take into account the entire life cycle of a product, not only from a functional and structural point of view, but also from its expected behavioral characteristics (maintainability, recyclability ...). | |
| − | ; | + | ; Data models and stochastic optimization |
| − | + | The stochastic optimization methods, in particular genetic algorithms, have proven their performance in many fields. We are particularly interested in issues that require to take into account large amount of structured knowledge. This is the case for example for route optimization of a fleet of medical vehicles where many constraints are involved (legal, practical needs for equipment for patients ...). In these cases, it is important that the domain knowledge, possibly expressed as an ontology, can be effectively used in the calculation of costs, but also crosses and mutations in genetic algorithms. | |
== Publications == | == Publications == | ||
| − | + | Please see the [http://icube-web.unistra.fr/papr/appli.php?author=beuvron&title=&team=toutes&annee1=&annee2=&display=rap+&nationalRank=toutes&project=tous list] of my publications in the ICube laboratory index. | |
| − | == | + | == Teaching == |
| − | * | + | * JAVA programming : INSA Strasbourg, (bachelor) |
| − | * | + | * UML and Object Oriented Programming : INSA Strasbourg, (bachelor) |
| − | * | + | * Relational databases : INSA Strasbourg, (bachelor) |
| − | * Interfaces | + | * Event programming, Graphical User Interfaces : INSA Strasbourg, (bachelor) |
| − | * | + | * Logics for AI : INSA Strasbourg, (master) |
| − | |||
| − | |||
== Contact == | == Contact == | ||
Revision as of 11:28, 3 March 2016
Research
In the generic field of knowledge representation, my work aims to define an hybrid inference formalisms based on description logics and rules, and suitable for the formulation of problems related to the design, configuration, diagnosis, or classification of systems.
- Semantic interpretation of high resolution satellite image.
To enable the identification of high-level structured objects (house, street ...), it is necessary to identify and merge segmented regions from image analysis. A geographic ontology describing not only the thematic objects but also their qualitative and quantitative spatial relationships is used to guide this interpretation.
- Hybrid representation of knowledge for innovative design.
The conceptual phase of innovative design is discussed in the context of the TRIZ theory for which an ontology of problem-solving concepts is proposed. The aim is to take into account the entire life cycle of a product, not only from a functional and structural point of view, but also from its expected behavioral characteristics (maintainability, recyclability ...).
- Data models and stochastic optimization
The stochastic optimization methods, in particular genetic algorithms, have proven their performance in many fields. We are particularly interested in issues that require to take into account large amount of structured knowledge. This is the case for example for route optimization of a fleet of medical vehicles where many constraints are involved (legal, practical needs for equipment for patients ...). In these cases, it is important that the domain knowledge, possibly expressed as an ontology, can be effectively used in the calculation of costs, but also crosses and mutations in genetic algorithms.
Publications
Please see the list of my publications in the ICube laboratory index.
Teaching
- JAVA programming : INSA Strasbourg, (bachelor)
- UML and Object Oriented Programming : INSA Strasbourg, (bachelor)
- Relational databases : INSA Strasbourg, (bachelor)
- Event programming, Graphical User Interfaces : INSA Strasbourg, (bachelor)
- Logics for AI : INSA Strasbourg, (master)
Contact
François de Bertrand de Beuvron ICube - UMR 7357 Pôle API Bd Sébastien Brant - BP 10413 67412 Illkirch Cedex Tel : +33 (0)3 68 85 43 96 Fax : +33 (0)3 68 85 44 55 courriel : debeuvron _AT_ unistra _POINT_ fr