Apparently, everyone knows that tf-idf stands for "term frequency-inverse document frequency". I had no idea, and the article didn't have time to include a link to <a href="http://en.wikipedia.org/wiki/Tf%E2%80%93idf" rel="nofollow">http://en.wikipedia.org/wiki/Tf%E2%80%93idf</a> or even type out the acronym.
Two remarks:<p>1. Don't 'earmuff' your stopwords, since you don't intend them to be rebound. An according guideline can be found here: <a href="http://dev.clojure.org/display/design/Library+Coding+Standards" rel="nofollow">http://dev.clojure.org/display/design/Library+Coding+Standar...</a><p>2. You could replace <i>(remove nil? (map db (tokenize raw-text)))</i> with <i>(keep db (tokenize raw-text))</i>