CSCE 5200 Information Retrieval and Web Search
Spring 2008
Assignment 3
Issued: 03/23/2008 Due: 04/10/2008


In this assignment, you will have to implement a Naive Bayes text
classifier, as discussed in class. To avoid zero counts, make sure
you implement some form of smoothing. In your implementation, assume
the following format for the input data:
   - all the training files are stored in a directory 
     called training/
   - all the test files are stored in a directory
     called test/
   - there is a file called cats.txt which includes a line for 
     each training file and for each test file, each file name
     with its corresponding category

For instance, assuming the following files:
   training/ 
       file1 file3 file4 file6
   test/
       file2 file5
The file cats.txt might include something like:
       file1 spam
       file2 no-spam
       file3 spam
       file4 spam
       file5 no-spam
       file6 spam

Test your implementation on the data set provided on the class 
webpage. Answer the following question:
  1. What is the classification accuracy for your text classifier?
  2. What are the 10 most dominant word features for each class 
     (in other words, what are the words that have the highest 
      P(w|c)*P(c) for each class)


Submission instructions:
   - write a README file including:
         * a detailed note about the functionality of the program, 
         * complete instructions on how to run it
         * answers to the questions above 
   - make sure you include your name in each program and in the README file.
   - make sure the program runs correctly on the CSP machines.
   - submit your assignment, including programs and README file by the
     due date using the 'project' program. 
     Class code is 5200s001, project HW3.