Machine Learning on Linked Data, a Position Paper

Abstract. The combination of linked data and machine learning is emerging as an interesting area of research. However, while both fields have seen an exponential growth in popularity in the past decade, their union has received relatively little attention. We suggest that the field is currently too complex and divergent to allow collaboration and to at-tract new researchers. What is needed is a simple perspective, based on unifying principles. Focusing solely on RDF, with all other semantic web technology as optional additions is an important first step. We hope that this view will provide a low-complexity outline of the field to entice new contributions, and to unify existing ones.