SecureScan AI framework boosts malware and phishing detection accuracy to 93%

Researchers have developed SecureScan, a three-layer AI framework designed to detect sophisticated malware and phishing attempts. This system combines logistic regression for classification, heuristic analysis for initial filtering, and the VirusTotal API for external threat intelligence. SecureScan achieved 93.1% accuracy on benchmark datasets, with a focus on minimizing false positives through calibrated thresholds and gray-zone logic. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Offers a lightweight, effective alternative to complex deep learning models for cybersecurity threat detection.

RANK_REASON Academic paper detailing a new AI-driven security framework.

Read on arXiv cs.CV →

• SecureScan
• VirusTotal API
• arXiv
• Logistic Regression

• paper
• safety