As a Data Analyst for Wikidata and WDQS, I would like to know what are the various large subgraphs in Wikidata and what are the benefits/losses of splitting them off from Wikidata. The aim is to identify large subgraphs along with those already known (scholarly articles, astronomical objects) and find out how often these subgraphs are queried. This can be estimated from:
- The subgraph sizes
- Connection of subgraphs to other subgraphs
- Number of queries that inquire of this subgraph
- Number of queries that span multiple subgraphs (estimation of how much federation load)