Decision Trees

CSC 261 - Artificial Intelligence - Weinman



Answer the following questions. Record your answers in your Reading Journal.
  1. In your own words, briefly (3-5 sentences) explain the concept of information entropy1 and why it is useful for learning decision trees.
  2. Briefly describe an example from your own inductive learning experience that exemplifies the fourth recursive case (i.e., no remaining attributes) for a learning a decision tree classifier (pp. 700-701).
  3. Identify the line(s) of DECISION-TREE-LEARNING that remains the most confusing to you. Briefly explain what you find confusing about it.

Footnotes:

1The use of the term entropy in this context is distinct from its use in thermodynamics. However, it is closely related to its use in statistical mechanics.