I am having trouble figuring out the formula for Brunet's index which can be found in this paper http://csbapp.uncw.edu/data/mscsis/annalspaper.aspx?v=6&i=1&p=11.

I tried to apply it to interviews with ~1000 words, however, the values generated seems to be very extreme around e-100. Is this statistic not valid for large text?

enter image description here

  • What is the meaning of N and V? Are you sure that this is really the correct definition (go back to the original publication, don't rely on the paper in the link)? I tried a google book search, but some relevant pages were hidden, therefore only comments, no answer. Mar 11, 2018 at 9:14
  • My appologies, N in this case is the number of word token within the text and V is the word type.
    – kkawabat
    Mar 11, 2018 at 18:42

1 Answer 1


I found out that the formula was actually BI = N^V^(-.165) which returned more reasonable values.

  • Please accept your own answer so this no longer comes up as unanswered. Thanks.
    – tripleee
    Apr 12, 2018 at 5:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.