Pointwise mutual information

Creator

Creator

Created

Created

2025 Jan 26 22:10

Editor

Editor

Edited

Edited

2025 Mar 14 11:53

Refs

Refs

Pointwise mutual information

Pointwise mutual information (PMI) $[-\infty, \infty]$

Negative PMI values are harder to interpret and evaluate; relatedness is easier to evaluate as opposed to un-relatedness

PMI(x_i, c_j) = \log\frac{p(w_i, c_j)}{p(w_i)\cdot p(c_j)}

Positive Pointwise mutual information (PPMI)

PPMI(x_i, c_j) = \max(PMI(x_i, c_j), 0)

Pointwise mutual information

In statistics, probability theory and information theory, pointwise mutual information (PMI),[1] or point mutual information, is a measure of association. It compares the probability of two events occurring together to what this probability would be if the events were independent.[2]

Pointwise mutual information

https://en.wikipedia.org/wiki/Pointwise_mutual_information

Recommendations

/////////