SPARQL for music: when melodies meet ontology

Introduction by OpenMethods Editor (Marinella Testori):

Knowledge Graphs have been already object of posts on our OpenMethods’ platform (see the contribution about ArCo-Architecture of Knowledge: https://openmethods.dariah.eu/2021/03/11/cultural-ontologies-the-arco-knowledge-graph/).

Along with art and architecture, another cultural heritage oriented field of application of such a graph is Musicology , as detailed
in the paper “Knowledge Graph Construction From MusicXML: An Empirical Investigation with SPARQL Anything”, by Ratta and Daga
(http://oro.open.ac.uk/85326/).

Developed in the context of the EU H2020 Polifonia project, the investigation deals with the potentialities of SPARQL Anything to extract musical features, both at metadata and symbolic levels, from MusicXML files.

The paper captures the procedure that has applied by starting from an overview about the application of ontologies to music, as well as of the so- called ‘façade-based’ approach to knowledge graphs, which is at the core of the SPARQL Anything software. Then, it moves to an illustration of the passages involved (i.e., melody extraction, N-grams extraction, N-grams analysis and exploitation
of the Music Notation Ontology). Finally, it provides some considerations regarding the result of the experiment in terms of effectiveness of the queries’ performance.

In conclusion, the authors highlight how further studies in the field may cast an increasingly brighter light on the application of semantic-oriented methods and techniques to computational musicology.

References

Ratta, Marco and Daga, Enrico Knowledge Graph Construction From MusicXML: An Empirical Investigation With SPARQL Anything. In: Proceedings of the Musical Heritage Knowledge Graphs Workshop (Presutti, Valentina; Buffa, Michel; Steels, Luc; Trubert, Jean-François; Daga, Enrico and Meroño Peñuela, Albert eds.), CEUR-WS. http://oro.open.ac.uk/85326/