Most security threats use email as an attack vector. Since the amount of traffic conveyed in this way is particularly large, it is necessary to use automated detection procedures that exploit machine learning (ML) algorithms. In this chapter, different detection strategies ranging from linear classifiers and Bayesian filters to more sophisticated solutions such as decision trees, logistic regression, and natural language processing (NLP) will be illustrated.
This chapter will cover the following topics:
- How to detect spam with Perceptrons
- Image spam detection with support vector machines (SVMs)
- Phishing detection with logistic regression and decision trees
- Spam detection with Naive Bayes
- Spam detection adopting NLP
