Real-time phishing uniform resource locator detection based on hybrid embedding transformer and retraining-free inferencing

Key Points

• Real-time URL detection significantly improves identifying phishing attempts, enhancing online safety.
• The hybrid embedding transformer shows over 90% accuracy in recognizing malicious URLs.
• Analysis combines machine learning techniques with a robust hybrid model for efficient detection.
• This study highlights the need for advanced detection methods in addressing rising cybersecurity threats.