Graph lenses over any data: the ConnectionLens experience - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Graph lenses over any data: the ConnectionLens experience

Résumé

Data integration is decades-old problem that takes many shapes, depending on the model of the integrated data sources, the integration model (if a single model is used), the expressivity of the features supported from each model, etc. Over several years, we have worked to integrate very het- erogeneous data, aiming to address the needs Non-Technical Users (NTUs), notably journalists. The choice we make is to integrate data of any model by migrating (transforming) it into a graph, consisting simply of labeled nodes and edges. Such graphs are much simpler than Property Graphs and more basic even than RDF graphs, since we do not require URI labels on internal nodes. This experience paper gives an overview of our research efforts towards querying, understanding, and exploring the resulting graphs. The contributions we brought are in the areas of data integration, graph exploration, graph querying, and go towards managing semistructured data lakes.
Fichier principal
Vignette du fichier
main.pdf (1.03 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04591897 , version 1 (29-05-2024)

Licence

Identifiants

  • HAL Id : hal-04591897 , version 1

Citer

Oana Balalau, Nelly Barret, Simon Ebel, Théo Galizzi, Ioana Manolescu, et al.. Graph lenses over any data: the ConnectionLens experience. ICDE 2024 - 40th IEEE International Conference on Data Engineering, May 2024, Utrecht, Netherlands. ⟨hal-04591897⟩
3 Consultations
11 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More