Research areas

SPARKS (Scalable and Pervasive softwARe and Knowledge Systems) research focuses on the organization, representation and distributed processing of knowledge, as well as the extraction of this knowledge from data and its semantic formalization. Particular attention is paid, on the one hand, to large-scale architectures and massive data, and on the other hand, to the design of human-centred, knowledge-centred, evolutionary and adaptive software systems.

The division combines four themes:

• Knowledge Extraction and Learning: developing methods and algorithms based on machine learning, data mining, and artificial intelligence to extract new and useful information and knowledge from data;
• FOrmalizing and Reasoning with Users and Models: understanding data (1) by proposing multidisciplinary approaches for multi-criteria analysis and modelling of information systems, user communities and their interactions, and (2) by reasoning on these models using graph-oriented Semantic Web approaches to propose new analysis tools and to create new functionalities and better management;
• Scalable Software Systems: adapting and composing systems, data and workflows at different scales, from the local loop to massive distribution;
• Computer Science and Biology: computer science is needed to push the frontiers of knowledge in biology using, for example, ontologies, data mining, knowledge extraction, modelling and simulation of dynamic biological systems, formal proofs of biological system behaviour, and more generally computer-assisted model-based reasoning. There are also countless bio-inspired techniques that have made important contributions to research, such as neuroscience- and genetics-inspired learning techniques.

These themes should be understood as four dimensions or orthogonal axes against which the work of team members is positioned, with each member's activity potentially covering each of them to a greater or lesser extent.

The SPARKS division includes two joint project teams (EPCs) with Inria:

• WIMMICS (Web-instrumented man-machine interactions, communities and semantics), led by Fabien Gandon, which aims to bridge formal and social semantics on the Web, with research areas covering graph-oriented knowledge representation, reasoning and operationalisation to model and support actors, actions and interactions in epistemic communities on the Web;
• MAASAI, led by Charles Bouveyron (mathematician at the Jean Alexandre Dieudonné laboratory and director of 3IA Côte d'Azur), composed of both mathematicians and computer scientists, working on artificial intelligence models and algorithms to propose innovative learning methodologies, addressing real-world problems, that are theoretically sound, scalable and tractable.