Naive Bayes spam filter

Subtask1-simple Naive Bayes spam filter

• Use very simple data preprocessing (like stopwords) so that the emails can be read into the Naive Bayes
• Write simple Naive Bayes multinomial classifier
• Classify the data
• Choose a baseline and compare our classifier against it

Subtask 2 - Improve the simple Naive Bayes spam filter

• Use some smart feature processing techniques (Word stemming and Weighted Frequency and Odds (WFO)) to improve the classification results
• Compare the classification results with and without these techniques

Subtask 3 - Calibration and Naive Bayes extension

• Adaboosting
• Unknown tokens
• Graham’s Method