Section: Partnerships and Cooperations

National Initiatives


ANR ELKER (2017–2020)

Participants : Nacira Abbas, Miguel Couceiro, Amedeo Napoli.

The objectives of the ELKER ANR Research Project (https://project.inria.fr/elker/) are to study, formalize, and implement the search for link keys in RDF data [2]. Link keys generalize database keys in two independent directions, as firstly they deal with RDF data and secondly they apply across two relation datasets. In this project, we study the discovery of link keys and reasoning with link keys, being based on the FCA formalism. The ELKER project relies on the competencies of the Orpailleur Team in FCA and pattern structure algorithms, and also in partition pattern structures which are related to the discovery of functional dependencies. This project involves the EPI Orpailleur at Inria Nancy Grand Est, the EPI MOEX at Inria Grenoble Rhône Alpes, and LIASD at Université Paris 8.

ANR PractiKPharma (2016–2020)

Participants : Miguel Couceiro, Adrien Coulet, Pierre Monnin, Amedeo Napoli, Yannick Toussaint.

PractiKPharma for “Practice-based evidences for actioning Knowledge in Pharmacogenomics” is an ANR research project (http://practikpharma.loria.fr/) about the validation of domain knowledge in pharmacogenomics. Pharmacogenomics is interested in understanding how genomic variations related to patients have an impact on drug responses. While most of the available knowledge in pharmacogenomics –state of the art knowledge– lies in the biomedical literature, with various levels of validation, an originality of PractiKPharma is to use Electronic Health Records (EHRs) to constitute cohorts of patients where to discover knowledge units. Indeed, these cohorts are mined for discovering potential pharmacogenomics patterns to be then validated w.r.t. literature knowledge for becoming actionable knowledge units. More precisely, firstly we have to discover pharmacogenomic patterns from the literature, and secondly we should confirm or moderate the interpretation and validation of these units by mining EHRs. Comparing knowledge patterns extracted from the literature with facts extracted from EHRs is a complex task depending on the EHR language –the literature is in English whereas EHRs are in French– and on knowledge level, as EHRs represent observations at the patient level whereas the literature is related to sets of patients. The PractiKPharma involves three other laboratories, namely LIRMM in Montpellier, SSPIM in St-Etienne, and CRC in Paris.

ANR AstroDeep (2019–2022)

Participants : Miguel Couceiro, Amedeo Napoli, Claire Theobald.

Astronomical surveys planned for the coming years will produce data that present analysis challenges not only because of their scale (hundreds of petabytes), but also by the complexity of the measurement challenges on very deep images (for instance subpercent-level measurement of colors or shapes on blended objects). New machine learning techniques appear very promising: once trained, they are very efficient and excel at extracting features from complex images. In the AstroDeep project, we aim at developing such machine learning techniques that can be applied directly on complex images without going through the traditional steps of astronomical image processing, that lose information at each stage. The developed techniques will help to leverage the observation capabilities of future surveys (LSST, Euclid, and WFIRST), and will allow a joint analysis of data.

The AstroDeep ANR Project involves three labs, namely APC Paris (“Astroparticules et Cosmologie Paris”), the Orpailleur Team at Inria Nancy Grand Est/LORIA, and “Département d'Astrophysique CEA Saclay”.

Inria Project Labs, Exploratory Research Actions, and Technological Development Actions

Participants : Guilherme Alves Da Silva, Alexandre Bazin, Miguel Couceiro, Nyoman Juniarta, Tatiana Makhalova, Amedeo Napoli, Laureline Nevin, Abdelkader Ouali, Claire Theobald, Georgios Zervakis.

HyAiAI (IPL 2019-2022)

Recent progress in Machine Learning (ML) and especially in Deep Learning has made ML present and prominent in a wide range of applications. However, current and efficient ML approaches rely on complex numerical models. Then, the decisions which are proposed may be accurate but cannot be easily explained to the layman, especially in some cases where complex and human-oriented decisions should be made, e.g. to get a loan or not, to obtain a chosen enrollment at university. The objectives of the HyAIAI IPL are to study the problem of making ML methods interpretable. For that, we will design hybrid ML approaches that combine state of the art numerical models (e.g. neural networks) with explainable symbolic models (e.g. pattern mining). More precisely, one goal is to integrate high level domain constraints into ML models, to provide model designers information on ill-performing parts of the model, and to give the layman/practitioner understandable explanations on the results of the ML model.

The HyAIAI IPL project involves seven Inria Teams, namely Lacodam in Rennes (project leader), Magnet and SequeL in Lille, Multispeech and Orpailleur in Nancy, and TAU in Saclay.

Ordem (ADT 2019-2020)

One of the outputs of the former Hybride ANR project was the Orphamine system which aims at information retrieval and diagnosis aid in the domain of “rare diseases”. The Orphamine system is based on domain knowledge, and in particular on medical ontologies such as ORDO (“Orphanet Rare Diseases Ontology”) and HPO (“Human Phenotype Ontology”). In this way, the objective of the “Ordem” ADT is to update Orphamine, in making the system more accessible and more open. This requires many developments for developing the connections with domain knowledge, graph mining methods for retrieving relevant units in knowledge graphs, actual visualization tools, pattern mining, statistical decision tools for decision making (in particular log-linear models), and as well text mining tools for analyzing expert queries and medical literature about rare diseases. Such developments are and will be carried out until the end of next year, for making the system robust and publicly accessible through a web interface.

HyGraMi (PRE Inria 2018-2020)

Finally, the so called “projet de recherche exploratoire” (PRE) HyGraMi for “Hybrid Graph Mining for the Design of New Antibacterials” is about the fight against resistance of bacteria to antibiotics. The objective of HyGraMi is to design a hybrid data mining system for discovering new antibacterial agents. This system should rely on a combination of numeric and symbolic classifiers, that will be guided by expert domain knowledge. The analysis and classification of the chemical structures is based on an interaction between symbolic methods e.g. graph mining techniques, and numerical supervised classifiers based on exact and approximate matching. This year we work on a method based on tree decomposition for performing feature selection and improving data lining of such complex molecular structures [49].