Supervised Learning

• Supervised learning
  • from X (input) → Y (output)
    • (x): email → (y): spam? (0/1) ⇒ application: spam filtering
    • ad, user info → click ⇒ online advertising
  • visualization
  • 1. Regression: predict a number from infinite possible outputs
    • give the algo the so-called right dataset → learning algo will learn to plot and fit the lines straight or polynomial (depends) → output of the predicted one
  • 2. Classification: predict categories from a small number of outputs (eg. benign, malignant type 1, malignant type 2, in tumor prediction)