2
$\begingroup$

According to the wiki-article, http://en.wikipedia.org/wiki/Tf%E2%80%93idf the term frequency of term $t$ in document $d$ is given as follows: $$ tf(t,d) = 0.5 + \frac{0.5 \times f(t,d)}{\max{\{f(w,d): w~\epsilon~d\}}} $$

This is the augmented way of computation. But what does $w$ stand for?

$\endgroup$
0
$\begingroup$

A document is modeled as a set $d$ of words. Since we're maximizing $f(w,d)$ over all $w\in d$, this means that $w$ ranges over all possible words contained in $d$.

In the future, please check a textbook before asking. We expect you to do some research on your own before asking, and this is a standard topic that's well-covered in standard explanations of tf-idf.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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