PulseAugur
EN
LIVE 05:52:22

New hybrid pipeline boosts phishing detection with DistilBERT and URL analysis

By PulseAugur Editorial · [2 sources] ·

Researchers have developed a novel hybrid pipeline designed to enhance phishing and threat classification. This system integrates multiple engines, including a URL analysis stack, a DistilBERT NLP classifier, and a threat-intelligence synchronizer. The pipeline achieves a high F1 score of 0.914 on a benchmark of over 10,000 emails, significantly improving real-phishing recall while minimizing false positives. AI

IMPACT This hybrid approach could lead to more robust and accurate threat detection systems, improving online security.

RANK_REASON The cluster contains a research paper detailing a new technical approach to phishing detection.

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

New hybrid pipeline boosts phishing detection with DistilBERT and URL analysis

COVERAGE [2]

  1. arXiv cs.CL TIER_1 English(EN) · Saifelden M. Ismail, Aser O. Ibrahim, Omar A. Mahmoud ·

    A Hybrid, Multi-Layered Pipeline for Phishing and Threat Classification: Independently Validated URL and NLP Engines with a Calibrated Multi-Channel Fusion Stage

    arXiv:2606.21690v2 Announce Type: replace-cross Abstract: Phishing is a multi-modal threat. We present a hybrid pipeline that scores each modality with its own engine and fuses the results. Three engines are built, deployed, and independently benchmarked: a four-stage URL stack (…

  2. arXiv cs.CL TIER_1 English(EN) · Omar A. Mahmoud ·

    A Hybrid, Multi-Layered Pipeline for Phishing and Threat Classification: Independently Validated URL and NLP Engines with a Calibrated Multi-Channel Fusion Stage

    Phishing is a multi-modal threat. We present a hybrid pipeline that scores each modality with its own engine and fuses the results. Three engines are built, deployed, and independently benchmarked: a four-stage URL stack (Domain Guard, lexical model, threat intelligence, and an a…

RELATED ENTITIES

RELATED TOPICS