Classification (CS229)

發表於

Classification examples

  • Email: spam (Y/N)?
  • Online Transcations: fraudulant (Y/N)?
  • Tumor: in pathology, is it malignant or benign?

$y\in\{0,1\}$

$0$ implies abscense / negative class; $1$ implies exist / positive class.

Linear regression’s output $h_{\theta}(x)$ can be larger than $1$ or smaller than $0$ even if all given data $y$ is in domain $[0,1]$. To get rid of this situation, we use logistic regression to make $0\le h_{\theta}(x)\le1$.