Meta2KG: An Embeddings-based Approach for Transforming Metadata to Knowledge Graphs

Nora Abdelmageed, Birgitta König-Ries

Keywords: Metadata Analysis, RDF, Ontology Matching, Ontology Population, Knowledge Graphs, Embeddings

TL;DR: Metadata XML files transformation to a unified KG using embeddings-based approach.

Abstract:

Metadata is used to describe data.
It includes information about the who, when, where, how, and why of data collection.
Ideally, it should be in a machine-understandable format like RDF.
This enables data queries using structured query languages like SPARQL and empowers further data usage.
In this paper, we investigate metadata as a source for generating Knowledge Graphs (KGs).
We introduce a semi-automated approach that transforms raw metadata files into a KG.
We develop the Biodiversity Metadata Ontology (BMO) as an underlying schema for our technique.
We auto-populate the constructed ontology with instances from several metadata files as a unified KG.
Finally, we discuss the common obstacles that face such a transformation procedure.
Our results show that metadata files are a promising source for KG construction.
In addition, our resources and code are publicly available (https://github.com/fusion-jena/Meta2KG).