Collocations are high mutual-information (high MI) phrases that are more frequent qua phrases than the average frequency of their component words separately. For ex, in some corpus Puerto Rico has appears 10 times, Puerto 11 times, and Rico 11 times. IE, these words appear mainly in this one phrase, not independently. The frequency of Puerto Rico is 10, averaged frequency of the component terms is (11+11)/2 = 10.5, so the ratio of phrase to individual word frequency is 10:10.5, or .95. MI becomes interesting at ratios of about .5 or .7 (see examples in the samples provided). Input Max 800k words.

NOTE: Calculations are made entirely from within the entry text itself, not using a list from outside the text. For a list-based phrase/collocation profiler see PHRASE PROFILER

(1) Upload TXT        (Only first 700k wds are used) OR Choose Big Text/Corpus ================ DONALD TRUMP (170k wds) BARACK OBAMA (200k wds) HILARY CLINTON (118k wds) BAWE PHYSICS (197k wds) BAWE METEOROLOGY (48k wds) BAWE PSYCHOLOGY (215k wds) BAWE CHEMISTRY (139k wds)

(2) Choose       M-I       level HIGH fewer/better .9 .7 .5 .3 .1 LOW more (≅ N-Gram)

(3) Click

(4) Get Result >