This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: CS221 Exercise Set #4 1 CS 221, Autumn 2006 Exercise Set #4 Handout #18 1. Decision Trees and Boolean Functions Give decision trees to represent the following boolean functions 1 : (a) A XOR B (b) [ A ∧ B ] ∨ [ C ∧ D ] 2. Decision trees and Presidential Elections Paul used to live in Palm Beach County, where he was hired as a consultant during the Florida recount controversy, in an attempt to leverage his machine learning expertise to the problem. Though his recommendations were never used (Paul was fired for...certain personal improprieties), Paul maintains that the key issue was whether or not a voter’s intended vote was counted and has created the following decision tree to determine this. The tree uses four binary features: B the B allot confuses you A you A sk for help understanding the ballot D you D imple the chad on your ballot (as opposed to punching it out) C the C ourts decide dimpled chads must be counted Assume that this tree correctly classifies everyone’s intended vote as positive (counted) or negative (not counted): 1 Problem taken from T. Mitchell, Machine Learning , McGraw Hill, 1997. CS221 Exercise Set #4 2 Suppose you don’t have Paul’s tree, but instead are given a large number of training instances (Florida voters), each classified consistently with this tree, which evenly and...
View Full Document
This note was uploaded on 11/30/2009 for the course CS 221 taught by Professor Koller,ng during the Winter '09 term at Stanford.
- Winter '09
- Artificial Intelligence