Simple approach: “gold standard” is a subset of a well-known dataset (obtained with a SPARQL CONSTRUCT query against the dataset's endpoint). Compute the set difference between, say, the labels of the resources in the gold standard and the labels of the resources in the dataset under assessment. For now, we'll do exact matches, not approximate matches.
Simple approach: “gold standard” is a subset of a well-known dataset (obtained with a SPARQL CONSTRUCT query against the dataset's endpoint). Compute the set difference between, say, the labels of the resources in the gold standard and the labels of the resources in the dataset under assessment. For now, we'll do exact matches, not approximate matches.