To get this to work you'll need the Stanford NLP engine files (e.g. stanford-corenlp-3.5.2.jar and stanford-corenlp-models-current.jar).
First you run import.rb to get the data in (make sure the Neo4j URL is correct) and then extract_authors.rb to extract authors from that data.