Hands-On Artificial Intelligence for Cybersecurity

Spam filters in action

How does an anti-spam algorithm actually behave in the classification of emails? First of all, let's classify the emails based on suspicious keywords. Let's imagine, for the sake of simplicity, that the list of the most representative suspicious keywords is thus reduced to only two words: buy and sex.

At this point, we will classify the email messages within a table, showing the number of occurrences of the individual keywords identified within the text of the emails, indicating the messages as spam or ham:

At this point, we will assign a score to every single email message.

This score will be calculated using a scoring ...