Information Gain in Decision Trees - Constructing A Decision Tree Using Information Gain

Constructing A Decision Tree Using Information Gain

A decision tree can be constructed top-down using the information gain in the following way:

  1. begin at the root node
  2. determine the attribute with the highest information gain which is not already used as an ancestor node
  3. add a child node for each possible value of that attribute
  4. attach all examples to the child node where the attribute values of the examples are identical to the attribute value attached to the node
  5. if all examples attached to the child node can be classified uniquely add that classification to that node and mark it as leaf node
  6. go back to step two if there are unused attributes left, otherwise add the classification of most of the examples attached to the child node

Read more about this topic:  Information Gain In Decision Trees

Famous quotes containing the words constructing, decision, tree, information and/or gain:

    The very hope of experimental philosophy, its expectation of constructing the sciences into a true philosophy of nature, is based on induction, or, if you please, the a priori presumption, that physical causation is universal; that the constitution of nature is written in its actual manifestations, and needs only to be deciphered by experimental and inductive research; that it is not a latent invisible writing, to be brought out by the magic of mental anticipation or metaphysical mediation.
    Chauncey Wright (1830–1875)

    There are many things children accept as “grown-up things” over when they have no control and for which they have no responsibility—for instance, weddings, having babies, buying houses, and driving cars. Parents who are separating really need to help their children put divorce on that grown-up list, so that children do not see themselves as the cause of their parents’ decision to live apart.
    Fred Rogers (20th century)

    On every tree a bucket with a lid,
    And on black ground a bear-skin rug of snow.
    The sparks made no attempt to be the moon.
    They were content to figure in the trees
    As Leo, Orion, and the Pleiades.
    And that was what the boughs were full of soon.
    Robert Frost (1874–1963)

    Theories of child development and guidelines for parents are not cast in stone. They are constantly changing and adapting to new information and new pressures. There is no “right” way, just as there are no magic incantations that will always painlessly resolve a child’s problems.
    Lawrence Kutner (20th century)

    We go to gain a little patch of ground
    That hath in it no profit but the name.
    William Shakespeare (1564–1616)