Information Gain in Decision Trees - General Definition

General Definition

In general terms, the expected information gain is the change in information entropy from a prior state to a state that takes some information as given:

Where denotes the information entropy.

Read more about this topic:  Information Gain In Decision Trees

Famous quotes related to general definition:

    The following general definition of an animal: a system of different organic molecules that have combined with one another, under the impulsion of a sensation similar to an obtuse and muffled sense of touch given to them by the creator of matter as a whole, until each one of them has found the most suitable position for its shape and comfort.
    Denis Diderot (1713–1784)