Machine Learning Algorithms - Second Edition

Completeness

A complementary requirement is that each sample belonging to a class is assigned to the same cluster. This measure can be determined by using the conditional entropy H(K|C), which is the uncertainty in determining the right cluster given the knowledge of the class. Like for the homogeneity score, we need to normalize this by using the entropy H(K):

The conditional entropy, H(K|C), and the entropy, H(K), can be computed using the frequency counts in the same way that was shown in the previous section.

We can compute this score (on the same dataset) by using the completeness_score() function:

from sklearn.metrics import completeness_score ...