CS430 Class Project: What to Turn in

We will test your spam filter by training it on a set of email messages and then testing it on additional email messages.

Please provide the following information when you turn in your spam filter:

• Source code for the program
• Makefile (or compilation instructions) for compiling the run-time executable.
• Any additional files that the program requires (e.g., files containing the counts you have collected during training)
• Instructions on how to run the program.
  • The program should be executable from the command line
  • The program should read an email message from standard input and either train on it (if it contains a special header, "X-SPAM-LABEL:") or classify it (if it does not).
• Instructions on how to set the false positive/false negative tradeoff parameter
• A 1-2 page description of the features that you are using in your spam filter.
• Email all of this information (with attachments for the various files) to Hongli Deng by 9:30am December 8.

The evaluation process:

• The program will be executed on approximately 1000 email messages.
• We will execute it using 2 different settings of your false positive/false negative tradeoff parameter.