Hybrid Expert
System for
Hybrid Expert System for
Phishing WebsiteDetection
Analyze suspicious URLs using forward chaining expert rules and optimized XGBoost machine learning.
F01-F30
Knowledge Base
91
ML Features
Hybrid
XGBoost Runtime
Phishing Intelligence
Risk signals explained before the model speaks
PhishGuard keeps the rule base visible so the detection process is explainable during demos, audits, and academic review.
What is Phishing?
Phishing websites impersonate trusted services to steal passwords, OTP codes, banking data, or personal information.
Why It Matters
A single fake login page can lead to account takeover, financial loss, or identity theft.
How PhishGuard Works
The system extracts F01-F30 facts, evaluates expert rules using forward chaining, then validates the result using optimized machine learning.
System Flow
Expert-system-first detection pipeline
The visual order mirrors the backend runtime: rules and forward chaining come before the optimized ML decision layer.
URL Input
User submits a website address for defensive analysis.
Feature Extraction
Runtime extracts symbolic facts F01-F30.
Forward Chaining
Expert rules evaluate working memory first.
XGBoost Prediction
Primary ML model validates the risk pattern.
Hybrid Decision
Final result combines inference and ML evidence.
Runtime mode: Optimized Hybrid XGBoost
Analyze a Website URL
Enter a URL to evaluate phishing risk using expert rules and machine learning.