A Comprehensive Analysis of Adversarial Attacks against Spam Filters

TL;DR

This paper thoroughly examines how adversarial attacks compromise deep learning-based spam filters, evaluating multiple models and attack types to identify vulnerabilities and suggest improvements for enhanced security.

Contribution

It provides a comprehensive evaluation of adversarial attack impacts on various deep learning spam filters and introduces novel scoring functions to assess attack effectiveness.

Findings

Deep learning models are vulnerable to adversarial attacks at multiple levels.

Novel scoring functions improve attack success measurement.

Analysis highlights key weaknesses in current spam filters.

Abstract

Deep learning has revolutionized email filtering, which is critical to protect users from cyber threats such as spam, malware, and phishing. However, the increasing sophistication of adversarial attacks poses a significant challenge to the effectiveness of these filters. This study investigates the impact of adversarial attacks on deep learning-based spam detection systems using real-world datasets. Six prominent deep learning models are evaluated on these datasets, analyzing attacks at the word, character sentence, and AI-generated paragraph-levels. Novel scoring functions, including spam weights and attention weights, are introduced to improve attack effectiveness. This comprehensive analysis sheds light on the vulnerabilities of spam filters and contributes to efforts to improve their security against evolving adversarial threats.