We should get rid of the "remove_duplicates" step. If the sparql query is formed well (aka one statement per property) there should be no duplicates for the same person.
Note: I decided to implement the duplicate-reconciliation in Python in a separate step (remove-duplicates.py and format-mdbs.py) instead of doing it directly in the SPARQL query.
We should get rid of the "remove_duplicates" step. If the sparql query is formed well (aka one statement per property) there should be no duplicates for the same person.