Section: Partnerships and Cooperations

International Initiatives

Inria Associate Teams:SNOWFLAKE


Participants : Adrien Coulet, Joël Legrand, Pierre Monnin, Malika Smaïl-Tabbone.

  • Title: Knowledge Discovery from Linked Data and Clinical Notes

  • International Partner (Institution - Laboratory - Researcher):

    • Stanford (United States) - Department of Medicine, Stanford Center for Biomedical Informatics Research (BMIR) - Nigam Shah

  • Start year: 2014

  • Web site: http://snowflake.loria.fr/

Snowflake is an Inria Associate Team which started in 2014. It is aimed at facilitating the collaboration between researchers from the Inria Orpailleur team and the Stanford Center for Biomedical Informatics Research, Stanford University, USA. The main objective of Snowflake is to improve biomedical knowledge discovery by connecting Electronic Health Records (EHRs) with domain knowledge either in the form of ontologies or of Linked Open Data (LOD). Such a connection should help to complete domain knowledge w.r.t. EHRs. The initial focus of Snowflake is the identification and characterization of groups of patients w.r.t. (adverse) reactions to drugs. Identified features associated with such groups of patients could be used as predictors of over- or under-reactions to some drugs.

Participation in Other International Programs

A stay at NASA Frontier Development Lab

Participant : Chedy Raïssi.

In 2013, NASA presented the “Asteroid Grand Challenge”, a White House supported initiative to supplement the NEO (Near-Earth Object) Program, with a mission: “Find all asteroid threats to human populations and to know what to do about them." There remain a number of unresolved gaps in this challenge, both in terms of discovery, characterization and eventual mitigation strategies, should a potentially hazardous asteroid (PHA) be discovered. By bringing new approaches in computer science, such as deep learning and data mining to tackle specific parts of the problem, solutions may be revealed that, combined with existing processes, significantly benefit the community as a whole.

In July 2016, Chedy Raïssi visited NASA Ames and SETI Institute as part of the Frontier Development Lab. He worked there on developing meaningful research opportunities, as well as support the work of the planetary defense community and show the potential of this kind of applied research methodology to deliver breakthrough of significant value. The work was over a period of six weeks, focusing on Delay-Doppler radar imaging. Delay-Doppler radar imaging is a powerful technique to characterize the trajectories, shapes, and spin states of near-Earth asteroids and has yielded detailed models of dozens of objects. Since the 1990s, Delay-Doppler data has been analyzed using the SHAPE software developed originally by Steven J. Ostro. SHAPE performs sequential single-parameter fitting, and requires considerable computation runtime and human intervention. Reconstructing asteroid shapes and spins from Delay-Doppler data is, like many inverse problems, computationally intensive and requires extensive human oversight of the shape modeling process. Chedy Raïssi has explored two new techniques to better automate Delay-Doppler shape modeling: Bayesian optimization and deep generative models.

Ciência Sem Fronteiras (2014–2016)

Participant : Amedeo Napoli.

Program “Ciência Sem Fronteiras” is a Brazilian research fellowship which provides a funding for the stay of a visiting French researcher in Brazil at Universidade Federal Pernambuco Recife for three years. The on-going project is called “Formal Concept Analysis as a Support for Knowledge Discovery” and is aimed at combining FCA methods with numerical clustering methods used by Brazilian colleagues. This project is supervised in Brazil by Professor Francisco de A.T. de Carvalho (CIn/UFPE).

The project aims at developing and comparing classification and clustering algorithms for complex data (especially interval and multi-valued data). Two families of algorithms are studied, namely “clustering algorithms” based on the use of a similarity or a distance for comparing the objects, and “classification algorithms in Formal Concept Analysis (FCA)” based on attribute sharing between objects. The objectives here are to combine the facilities of both families of algorithms for improving the potential of each family in dealing with more complex and voluminous datasets.

STIC AmSud: Autonomic Knowledge Discovery (AKD, 2015–2016)

Participants : Miguel Couceiro, Esther Galbrun, Amedeo Napoli, Chedy Raïssi.

This research project involves researchers with different specialties, from Brazil (Universidade Federal Rio Grande do Sul), from Chile (UFSM Santiago and Valparaiso), from Uruguay (Universidad de la Repùblica), and the Orpailleur Team. The project is interested in the design of solutions able to proactively understand the behavior of systems and networks in order to prevent vulnerable states. Accordingly, we aim at integrating knowledge discovery techniques within autonomic systems in order to provide intelligent self-configuration and self-protection mechanisms. The results of this project may not only benefit to end-users but also highly contribute to the scientific community by providing solid foundations for the development of more secure, scalable, and reliable management approaches.


Participant : Jean-Sébastien Sereni.

LEA STRUCO is an “Associated International Laboratory” of CNRS between IÚUK, Prague, and LIAFA, Paris. It focuses on high-level study of fundamental combinatorial objects, with a particular emphasis on comprehending and disseminating the state-of-the-art theories and techniques developed. The obtained insights shall be applied to obtain new results on existing problems as well as to identify directions and questions for future work. Jean-Sébastien Sereni is the contact person for LEA STRUCO which was initiated when Jean-Sébastien was a member of LIAFA.

Research Collaboration with HSE Moscow

Participants : Miguel Couceiro, Adrien Coulet, Amedeo Napoli, Chedy Raïssi, Justine Reynaud.

An on-going collaboration involves the Orpailleur team and Sergei O. Kuznetsov at Higher School of Economics in Moscow (HSE). Amedeo Napoli visited HSE laboratory several times while Sergei O. Kuznetsov visited Inria Nancy Grand Est several times too. The collaboration is materialized by the joint supervision of students (such as for example the thesis of Aleksey Buzmakov defended at the end of 2015), and the the organization of scientific events, as in particular the workshop FCA4AI whose fifth edition was organized this year in August at ECAI 2016 (see http://www.fca4ai.hse.ru) [53]. A special session about Knowledge Discovery and Formal Concept Analysis will be supervised by Sergei O. Kuznetsov and Amedeo Napoli at the next ISMIS Conference in Warsaw (Poland) next June 2017 (http://ismis2017.ii.pw.edu.pl/s_kd_fca.php).