Accurator is the demonstrator of the process of enrichment of Cultural Heritage data, specifically that of the Rijksmuseum. A live demo can be seen on http://rma-accurator.appspot.com.
There are two datasets available that were gathered during two experiments.
Dataset 1 is described in the paper Crowd vs. experts: nichesourcing for knowledge intensive tasks in cultural heritage. The experiment was performed in June 2013. The dataset contains the annotation from 4 experts and 75 crowd workers. The dataset can be found here.
Dataset 2 is described in the paper Crowdsourcing Knowledge-Intensive Tasks in Cultural Heritage. More information about the paper and the dataset can be found here.