# A2 - Assignment 2: Bayesian Networks and Classifiers...

This preview shows pages 1–3. Sign up to view the full content.

Assignment 2: Bayesian Networks and Classifiers CSC384H – Fall 2009 Out: October 7, 2009 Due: November 2, 1 p. m. Written work should be handed in during class. Programming should be handed in via CDF, before class. Assignment is worth 15% of your final mark 1. Bayesian Network Inference (4 points). In the Bayesian Network shown below, all variables are binary with values Dom(A)={A,~A}, Dom(B)={B,~B}, etc. The probability table values are as follows: P(A)=0.9 P(B)=0.7 P(G)=0.75 P(D | C) = 0.9 , P(D | ~ C) = 0.06 P(C | A,B) = 0.95 , P(C | A,~B) = 0.01 , P(C | ~A,B) = 0.6 , P(C | ~A,~B) = 0.03 P(E | C,G) = 0.95 , P(E | C,~G) = 0.01 , P(E | ~C,G) = 0.9 , P(E | ~C,~G) = 0.3 P(F | C,G) = 0.8 , P(F | C,~G) = 1.0 , P(F | ~C,G) = 0.1 , P(F | ~C,~G) = 0.5 Compute the following probabilities (show your work): a) P(E | A) (1/2 point) b) P(D | A,B) (1/2 point) c) P(F | ~E) (1.5 points) d) P(G | B,F) (1.5 points)

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2. (5 points, 1 point each) For the Bayesian Network below, answer the following questions. Justify your answers using the d-separation criterion. a) List all nodes dependent on 4. b) List all nodes dependent on 4 given 11. c) List all nodes dependent on 1 given 5. d) List all nodes dependent on 7 given 9. e) List all nodes dependent on 2 given 8. 3 (6 points, 5 for answers to questions and 1 for clear, concise and correct code). In this part of the assignment, you will implement a naive bayes classifier and apply it to stock data. Your specific goal will be to predict stock market trends in the days following the appearance of a semi-annual report, given various features of those semi-annual reports. The tool you will be using to implement your classifier will be MATLAB.
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 11/01/2010 for the course COMPUTER S CSC384 taught by Professor Sheila during the Fall '05 term at University of Toronto.

### Page1 / 4

A2 - Assignment 2: Bayesian Networks and Classifiers...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online