Dylan Hunter

For the mathematically-inclined, (or at least those not frightened off by! ) if you have a fairly active forum, and want to start getting a better sense for which concepts / topics are discussed more and less often across threads, you can use this sort of analysis to help surface these, for use for further product development, customer acquisition, etc.

(Note: it talks about comparing a query to a document, but since there’s no conceptual difference between a query and a document - they’re both just bundles of words - can also be used to compare any document in a collection to any other document in that collection).

Just found a mighty fine (MS Word) macro that when you run on any given document gives you a list of every unique word in the document, (minus any words you don’t care about, like “the,” “and,” “or,” etc) with how frequently it occurred. Mighty fine job, Allen!