Abstract
Music is a key component of our cultural heritage as well as a driver for the creative industry. As such it is studied from a social science and humanities perspective as well as from a computer science one.
Recently, knowledge graphs have shown the potential to become an enabling technology for the hybridisation and collaboration between these two research worlds. They can be used for feeding AI applications to support musicologists, musicians, producers, etc. in tasks such as music management, analysis, composition, and generation. And they are key for developing a new generation of music information retrieval applications, based on interlinked knowledge as opposed to isolated collections.
This workshop intends to bring together an interdisciplinary audience of researchers and practitioners, including digital artists, to present their most recent use cases and results on methods, tools and applications for building, analyzing, exploiting, and interacting with musical heritage knowledge graphs.
Motivation
Current music data available on the web and in institutional collections is served in heterogeneous and incompatible formats that cannot be easily integrated.
Knowledge graphs and semantic web technologies are now under the lens of this community as a possible enabler of a paradigm shift. Evidence of the timely character of this topic is also provided by numerous research projects that have been developed in recent years (DOREMUS, WASABI, Polifonia, DigThatLick - to name a few). Music streaming providers are investing in AI research to develop services and functions that rely on semantic analysis of music content.
Similarly, cultural and memory institutions and music collection owners are looking more and more at LOD-based solutions to improve data integration strategies and automatic analysis of their content towards better procedures for preservation, management, and enhancement of musical heritage. The Semantic Web community can respond to this call by putting into action focused research on this domain.