The ProQuest platform uses the relevance ranking system that is built into the FAST search engine.
A variety of factors help to determine the relevance of any particular document according to the query entered by the user. For example, the number of times a search term appears is factored in.
And in the case of multiple term queries, the proximity of the search terms to one another within any given field of metadata is factored in as well. ProQuest leverages more structured metadata than the average Web search engine—allowing us to weight various metadata fields against one another. For example, we give a higher relevance weighting to professionally-applied indexing fields like subjects, locations, and other points of indexing over other parts of the document, such as the full-text and tags added by users.