Skip to content
Exit presentation

Three Kinds of Guess, and Their Data

Named group, number or found groups?

One program names leaves. One guesses how much rain. One groups shells. Which is which?

Labelled leaves go in. A named group comes out.
Step 1 of 4
  1. Labelled leaves go in. A named group comes out.
  2. Past rainfall goes in. A number comes out.
  3. Unmarked shells go in. Groups it found come out.
  4. What comes out names the job.

Naming leaves is classification; guessing rain is regression; grouping shells is clustering.

Classification sorts things into named groups. It learns from labelled examples. A labelled example comes with its answer already on it. Regression guesses a number. Clustering finds groups with no labels. Unmarked shells have no labels.

Three kinds of guess side by side
KindLearns fromWhat comes out
ClassificationLabelled examplesA named group
RegressionPast examples with the answer knownA number, like a price
ClusteringNothing labelledGroups it found

Ask what comes out. A named group is classification. A number to guess, like a price, is regression. Groups found with no labels are clustering.

Watch out!

A yes or no answer is a named group, not a number. Two groups is still classification. Digits can still be names, like a class number.

Read each job. Pick the kind of guess.

Loading lab…

Jump to lesson

This track

  1. 0 Start here
  2. 1 Guessing a number from past data
  3. 2 Making groups when nobody gave labels
  4. 3 Named group, number or found groups?
  5. 4 What a dataset holds
  6. 5 Structured, unstructured or both?
  7. 6 Train, check, test
  8. 7 The big picture
  9. 8 Track quiz

Other tracks