New MLP framework detects phishing URLs with 99.24% accuracy

Researchers have developed a new, efficient framework for detecting phishing URLs in real-time. This system uses a hybrid approach, combining blacklist screening with a Multi-Layer Perceptron (MLP) classifier that analyzes structural URL features. It achieves high accuracy and speed by avoiding complex content analysis or external API calls, making it suitable for resource-limited environments. A prototype application called CyberGuard demonstrates its practical deployment potential. AI

IMPACT This research offers a more efficient and accurate method for real-time phishing URL detection, potentially improving cybersecurity defenses.

RANK_REASON This is a research paper detailing a new framework for phishing URL detection. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• CyberGuard
• PhiUSIIL

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →