Data Cleaning publication categorizer

Schema-level heterogeneity represents an obstacle for auto-
mated discovery of coreference resolution links between in-
dividuals. Although there is a multitude of existing schema
matching solutions, the Linked Data environment diers
from the standard scenario assumed by these tools. In par-
ticular, large volumes of data are available, and repositories
are connected into a graph by instance-level mappings. In
this paper we describe how these features can be utilised to
produce schema-level mappings which facilitate the instance
coreference resolution process. Initial experiments applying
this approach to public datasets have produced encouraging
results.