Annonce en lien avec l’Action/le Réseau : aucun

Laboratoire/Entreprise : Lamsade, Université Paris-Dauphine
Durée : 4 à 6 mois
Contact : kbelhajj@googlemail.com
Date limite de publication : 2020-06-01

Contexte :
Knowledge graphs can be viewed as large collections of interconnected entities enriched with semantic annotations. They have become powerful assets for enhancing search and are now widely used in both academia and industry. Well-known knowledge graphs include Google’s Knowledge Graph, Facebook’s Graph Search and Yago (see (Pan et al. 2017)). In the context of this internship, the focus will be on knowledge graphs that are available or can be exported as RDF datasets (Manola and Miller 2004) enhanced with RDF Schema (Guha and Brickley 2014) statements that capture relevant domain background knowledge.

Sujet :
By and large, available knowledge graphs lack provenance support (Cheney, Chiticariu, and Tan 2009). Provenance information informs on the how-about of entity, i.e., how they come to be, and can be used in a range of applications, e.g., to explain the results of a query or search, to propagate annotations among the enti- ties that constitute the knowledge graph, to learn attribution information, to name a few. The main objective of the internship will be to investigate how RDF knowledge graphs can be enhanced with provenance support. In doing so, the candidate will examine the different kinds of provenance information that can be collected and recorded and the computational complexity incurred by each, design an algorithmic solution, and implement it. For validation purposes, we will be using real knowledge graphs that are freely available under the aegis of the open linked data initiative.

Profil du candidat :
A master or Engineer student in computer science.

Formation et compétences requises :
– Semantic Web (RDF, RDFS)
– Python

Adresse d’emploi :
Université Paris-Dauphine, Place du Maréchal de Lattre de Tassigny, 75016 Paris

Document attaché : Master-Internship-1.pdf