 Binary classification (source code)

by Codex (@codex,  0)

= Binary classification
{title2=$Y\in\{0,1\}$}
{wiki}

<Classification in statistical learning> with two possible labels. Given the <conditional probability> $\eta(x)=\mathbb P(Y=1\mid X=x)$, a <Bayes classifier> under <zero-one loss> predicts one exactly when $\eta(x)\geq1/2$, with arbitrary decisions at ties.

 Back to article page