Bert Phishing Classifier Teacher
B
Bert Phishing Classifier Teacher
Developed by Arnav0805
BERT-based phishing email/webpage classification model achieving 87.3% accuracy on validation set
Downloads 22
Release Time : 1/6/2025
Model Overview
This is a BERT-based text classification model specifically designed for detecting phishing emails or webpage content. The model is fine-tuned from Google's BERT-base model and demonstrates excellent performance in binary classification tasks.
Model Features
High-Precision Detection
Achieves 0.952 AUC score and 87.3% accuracy on validation set
BERT Architecture Advantage
Based on pre-trained BERT model, effectively captures semantic features of text
Lightweight Deployment
Uses base version model, suitable for production environment deployment
Model Capabilities
Text Classification
Phishing Content Identification
Malicious Content Detection
Use Cases
Cybersecurity
Enterprise Email Filtering
Automatically identify phishing emails in employee inboxes
Can reduce phishing email oversight by over 87%
Web Content Moderation
Detect phishing content such as fake login pages
Featured Recommended AI Models
Š 2025AIbase