Machine learning for email spam filtering: review, approaches and open research problems.

Area of Science:

• Computer Science
• Information Security

Background:

• Increasing volume of spam emails necessitates robust antispam solutions.
• Machine learning (ML) is widely adopted for effective spam detection by major internet service providers (ISPs).

Purpose of the Study:

• To systematically review popular machine learning-based email spam filtering approaches.
• To analyze current research trends, efficiency, and challenges in spam filtering.

Main Methods:

• Systematic literature review of machine learning techniques for email spam filtering.
• Comparative analysis of existing ML approaches, including their strengths and weaknesses.

Main Results:

• Machine learning methods show success in detecting and filtering spam emails.
• Identified open research problems and compared the efficiency of various ML algorithms.

Conclusions:

• Deep learning and deep adversarial learning are recommended as future techniques for enhanced spam filtering.
• These advanced ML methods show promise in effectively combating the growing spam email menace.