mutual information

Dependence between variables

Mutual information is a measure from information theory quantifying how much knowing one variable tells you about another.

If two variables are independent, their mutual information is zero; if one perfectly predicts the other, it's maximal.

The concept appears in statistics, machine learning, and cryptography wherever relationships between variables matter: We calculated the mutual information between gene expression and disease outcome.

applications

machine learning

Algorithms that improve from data

word association · neural networks · clustering · gene expression · bioinformatics · signal processing · speech recognition · text mining · decision trees

theory

Shannon

Irish river or name

data compression · divergence · uncertainty · random variables · expected value · noise

statistics

correlation

Statistical relationship measure

chi-square · association · bivariate · multivariate · statistical significance · hypothesis testing · stochastic

Categories
Opposites
statistical dependence

mutual informationorthogonality

State of being independent

noise · randomness · entropy · independence

type of relationship measured

mutual informationcausation

Production of effects

confounding

Family

information, informed, mutual, misinformation, uninformed, informative, informing, informs, mutually

Familiarity
Top 1,000,000
Profile
Used worldwide · Timeless · Abstract · Rated G