📊 Evaluation Module

Model Performance Research

Comparative evaluation of Rule-Based, Logistic Regression, Random Forest, and Hybrid models trained on 2,100 synthetic social media profiles.

Loading evaluation data…

ROC Curves

Receiver Operating Characteristic – AUC comparison

Feature Importances

Random Forest feature contribution weights

Confusion Matrix — Logistic Regression

Confusion Matrix — Random Forest

Model Metrics Comparison

Accuracy · Precision · Recall · F1 · AUC side-by-side

🏗️ System Architecture

Profile Input
▼
Feature Engineering
▼
Rule Engine
12 heuristic rules
+
ML Models
LR · Random Forest
▼
Hybrid Scorer
0.6 × Rule + 0.4 × RF
▼
Risk Score + Status Badge + AI Explanation