Risk Fusion MLP — Fuzzy Logic
Component 10b | Final Year Project
Fuses GNN + CVE + identity scores into risk score (0-10) with adaptive fuzzy categories. Trained on 2000 real NVD CVEs. Boundaries auto-calibrated from score distribution.
Quick Start
from risk_fusion import RiskFusionPipeline
pipe = RiskFusionPipeline.from_hub("sohomn/risk-fusion-mlp-fuzzy")
result = pipe.predict(structural=0.78, temporal=0.82, cvss=9.8,
exploitability=3.9, impact=5.9, identity_anomaly=0.65)
# Returns: Category + Certainty + Fuzzy Memberships + Alert Priority
Fuzzy Output
Final Risk Score: 8.9
Category: Critical
Certainty: 0.92
Fuzzy Memberships: {Critical: 0.92, High: 0.14}
Defuzzified: 8.71
Alert Level: 8.19
Alert Priority: IMMEDIATE
Performance
| Metric | Value |
|---|---|
| Test MAE | 0.0853 |
| Test RMSE | 0.1076 |
| Test R2 | 0.9921 |
License
MIT - Research use only.