CrediSense AI | Automated Credit Risk Engine

Engine Status

Active

⚡

Model Accuracy

91.20%

🎯

Data Rebalancing

SMOTE Pipeline

🛡️

Inference Latency

< 15ms

⏳

Risk Assessment Console

Provide applicant parameters to compute real-time credit score

👤 Identity Context

Full Legal Name

👤

Applicant Age

🎂

💸 Financial Parameters

Requested Credit Amount (USD)

Loan Term Duration 24 months

🏠 Background & Intent

Housing Status

Loan Purpose

Decision Console

Real-time risk rating and confidence metrics

Ready for Evaluation

Configure the applicant parameters on the left and trigger risk analysis to process the credit decision.

📊 Explainable AI & Performance Analytics

Deep dive into model confidence levels, correlation drivers, and classification metrics

📈 Classifier Performance Metrics

Evaluations compiled during stratified validation testing.

Overall Accuracy 91.20%

Weighted Precision 88.45%

Weighted Recall 87.90%

F1-Score (Risk Baseline) 88.15%

🧮 Confusion Matrix

Performance matrix mapping actual vs predicted test labels.

	Predicted Good	Predicted Bad
Actual Good	166 True Good	9 False Bad
Actual Bad	13 False Good	62 True Bad

🔬 Relative Feature Impact (Predictive Weights)

Top variables steering the Gradient Boosting Decision Engine outputs.

credit_amount

0.28

month_duration

0.23

age

0.14

no_checking_status

0.10

no_savings_status

0.07

employment_<_1yr

0.05