ML Operations

Model performance monitoring and training management

Isolation Forest

Unsupervised • v1.2
ACTIVE
Accuracy
92.3%
F1 Score
0.904
Inference
0.8ms

Random Forest

Supervised • v2.0
ACTIVE
Accuracy
97.1%
F1 Score
0.971
Inference
1.2ms

Autoencoder

Deep Learning • v1.1
ACTIVE
Accuracy
94.8%
F1 Score
0.941
Inference
3.5ms

Ensemble Configuration

Final Score = 0.30 × IF_score + 0.45 × RF_confidence + 0.25 × AE_error

Datasets: NSL-KDD (primary) • CICIDS2017 (validation)