Answers are grounded in this paper and cite their sources.
A number of “normalized” measures of entropy have been obtained to measure the “intrinsic” uncertainty of a probability distribution. Their graphs have been drawn and it has been shown that the normalized measures of a given probability distribution are much closer to one another than the corresponding absolute measures of entropy.
No takes yet. Share an insight, caveat, or question.
Kumar et al. (1986) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: