ℹ️ Upload any CSV file — with or without a Class column.
Accepts both raw transaction data and
PCA-transformed data (V1…Vn columns).
If Class column is present → model trains and evaluates. If no Class column → pre-trained model predicts each row.
If Class column is present → model trains and evaluates. If no Class column → pre-trained model predicts each row.
Drop your CSV file here
With Class column → full ML training + evaluation
Without Class column → instant predictions using trained model
✓file.csv
Running ML pipeline — this may take 30–90 seconds
Uploading file…
Overview
Model Comparison
| Model | Precision | Recall | F1-Score | ROC-AUC |
|---|
Transaction Predictions
| # | Actual | Predicted | Fraud Probability | Result |
|---|
Visualizations
Class Distribution
ROC Curves
Confusion Matrix
Precision / Recall / F1
Feature Importance
Enter a single raw transaction manually or upload a
single-row CSV. The app automatically applies
StandardScaler + PCA and predicts
Fraud or
Legitimate.
No Class column needed!
⚠️ Important: Run Batch CSV Analysis first
with a labeled dataset to train the model before using single transaction check.
Raw Transaction Details
Upload Single Transaction CSV
One row of raw transaction data
No Class column needed — model predicts automatically
✓file.csv
Analyzing transaction…