Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : Inria
Durée : 12 months
Contact : franck.michel@cnrs.fr
Date limite de publication : 2020-12-31

Contexte :
ISSA is a project funded by the Collex-Persée call for project, that involves three research institutes: Cirad, Inria and IMT Mines Alès.

Due to start in Oct. 2020, the project aims to improve access to and interoperability of the resources made available by scientific and technical information services, while offering innovative services meant for documentalists and researchers in a multidisciplinary and open science mindset. It seeks the provisioning of a generic solution leveraging interoperable metadata extracted from documentary resources. In this context, the goals of this project are twofold:
– Allow automatic indexing of documentary resources with thematic and geographic keywords from terminological resources (in the Semantic Web format) suitable for each domain or community;
– Demonstrate the interest of this approach by developing innovative search and visualization services intended for users, capable of exploiting this semantic indexing.

Agritrop, Cirad’s open publications archive (http://agritrop.cirad.fr), will serve as a use case and proof of concept throughout the project. The terminology resources will primarily be the Agrovoc thesaurus, Wikidata and GeoNames.

Sujet :
The recruited person will have a structuring role and a transversal activity. He/she will be in charge of designing and setting up an automated pipeline for the semantic indexing of a large corpus of scientific literature. This pipeline will notably rely on tools from the Science-Miner company (Grobid, entity-fishing). The recruited person will also apply this pipeline to the concrete case of the Agritrop scientific archive that consists of scientific articles but also other types of documents like maps.

The recruited person will take part to the reflection on the terminological resources used in the project, and to the definition and development of tools meant to exploit the semantic index: advanced search interfaces, geographical visualisations, enriched document visualization. These activities will involve the co-supervision of master trainees.

The recruited person will join the Inria center of Sophia Antipolis (France) as a research engineer, and will be working closely with the Wimmics Inria team, as well as remotely with the other partners of the project.

Main activities:
– Study existing tools for the automatic extraction and disambiguation of named entities against a knowledge graph, in particular the tools of Science-Miner
– Design and set up of an automated pipeline for the semantic indexing of scientific archive
– Deploy the pipeline for Agritrop, Cirad’s scientific archive
– Engage in a reflection about the terminological resources used in the project
– Write documentation and reports

Additional activities:
– Co-supervision of master trainees
– Participate in user training
– Present the work’ progress to partners

Profil du candidat :
The candidate must hold an engineering or master degree in Computer science with a 2-year experience, or a Ph.D in Computer science.

Formation et compétences requises :
The candidate shall have a strong expertise in working with knowledge graphs and Semantic Web technologies. He/she shall be very much at ease with common linux administration tasks. A solid experience in software development, notably Web development including common Javascript frameworks and REST APIs shall be a strong asset. Expertise in machine learning and text mining would also be appreciated.

Furthermore, the candidate will demonstrate aptitudes or matches with most of the following aspects:
– High motivation for working in a dynamic scientific research context
– Autonomy, remote working capabilities with collaborative tools
– Ability to collaborate with others on a common project
– Initiative, aptitude to propose technical solutions within the project goals
– Perfect English oral and writing skills

Adresse d’emploi :
Inria, Sophia antipolis, France

APPLICATIONS MUST BE SUBMITTED EXCLUSIVELY AT https://jobs.inria.fr/public/classic/fr/offres/2020-02901