Combining Text and Structural Features for Phishing

Why pure text features miss structural signals, a real feature-importance chart showing sender-link mismatch and urgency-word count outranking raw word counts, the Business Email Compromise case study where there is no malware payload at all, and the five-step combined feature pipeline, plus feature leakage.

~5 min total·4 quadrants of structured content

By the end of this topic, you will

Name four structural email features that combining with text features consistently outperforms either family alone
Summarise the Business Email Compromise case study and why it argues for structural, not just textual, features
Explain feature leakage with an original example and why it makes a model fail in production despite looking excellent in testing

Quadrant 1 · e-Tutorial

Video lectures and walkthroughs

Video content coming soon.

Quadrant 2 · e-Content

Articles and case studies

Combining Text and Structural Features for Phishing

The FBI's IC3 reported roughly 2.9 billion dollars in business email compromise losses in 2023, often with no malicious attachment at all. Structural features, sender-link mismatches and urgency language, matter as much as any file-based signal.

Quadrant 3 · Web Resources