KGSaw

One Size Does Not Fit All- Planning Methods for Data Fragmentation for Efficiently Creating Knowledge Graphs

verfasst von

Enrique Iglesias, Maria Esther Vidal

Abstract

This research addresses the challenges in planning knowledge graph (KG) creation. It presents KGSaw for partitioning and integrating data sources leveraging functional dependencies, while minimizing memory usage and execution time. Experimental results, involving existing KG creation engines, demonstrate KGSaw ability to enhance efficiency, with memory reductions up to 121.34 times and execution time improvements by a factor of 84.59. This emphasizes the importance of considering data source characteristics, like functional dependencies, in KG creation planning.

Organisationseinheit(en)

Institut für Data Science
Forschungszentrum L3S

Externe Organisation(en)

Technische Informationsbibliothek (TIB) Leibniz-Informationszentrum Technik und Naturwissenschaften und Universitätsbibliothek

Typ

Aufsatz in Konferenzband

Seiten

1668-1670

Anzahl der Seiten

3

Publikationsdatum

21.05.2024

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Software

Elektronische Version(en)

https://doi.org/10.1145/3605098.3636186 (Zugang: Geschlossen)