Extracts a table containing one row per document source in the database. Typically document sources are journals, but they can also be books, conference proceedings, or other types of publications. See Load ISI File Into Database/PERSON (XXX) for more information.
In addition to the standard Source fields in the database, 'Extract Document Sources' creates the aggregate column 'NUM_PAPERS_COUNTED_FROM_DATASET', which is the number of documents (usually papers) from the original dataset which are published in the document source.
This algorithm will only be as accurate as the journal disambiguation performed on the dataset. See Merge Document Sources.