Tekhnika/fraud-model-starter-pack-free

---
pretty_name: Fraud Model Starter Pack
language:
- en
license: cc-by-nc-4.0
tags:
- tabular-data
- analytics
- fintech
- fraud-detection
task_categories:
- tabular-classification
- time-series-forecasting
size_categories:
- 1K<n<10K
---

# Fraud Model Starter Pack

Free sample for fraud monitoring dashboards, suspicious activity analysis, and finance-oriented analytics workflows.

## What is included
- card_transactions.csv: 4398 rows, 10 columns
- cardholders.csv: 439 rows, 9 columns
- cards.csv: 593 rows, 9 columns
- daily_fraud_metrics.csv: 733 rows, 10 columns
- fraud_cases.csv: 967 rows, 10 columns
- merchant_risk_profiles.csv: 366 rows, 9 columns

## Why this dataset is useful
- Useful for a first fraud dashboard or suspicious activity notebook.
- Works well for SQL, notebooks, and BI prototyping.
- Provides a reduced but representative sample of the core workflow in the full starter pack.

## Starter use cases
- Fraud baseline using linked workflow and event data.
- Fraud monitoring dashboard for suspicious activity patterns.

## Schema overview
### card_transactions.csv
- Rows: 4398
- Columns: transaction_id, cardholder_id, card_id, merchant_id, transaction_date, merchant_category, transaction_amount, merchant_country, entry_mode, transaction_currency

### cardholders.csv
- Rows: 439
- Columns: cardholder_id, credit_score_band, customer_name, home_region, customer_segment, tenure_months, travel_frequency, historical_chargeback_flag, risk_band

### cards.csv
- Rows: 593
- Columns: card_id, cardholder_id, issue_date, status, card_product, network, credit_limit, card_present_usage_ratio, digital_wallet_flag

### daily_fraud_metrics.csv
- Rows: 733
- Columns: metric_id, merchant_id, metric_date, transactions_total, approved_amount, decline_rate, fraud_alerts_total, confirmed_fraud_total, chargeback_rate, loss_amount

### fraud_cases.csv
- Rows: 967
- Columns: fraud_case_id, transaction_id, cardholder_id, card_id, alert_date, fraud_type, investigation_status, chargeback_amount, loss_amount, model_score

### merchant_risk_profiles.csv
- Rows: 366
- Columns: merchant_id, merchant_category, merchant_country, merchant_name, acquirer_region, merchant_size, card_not_present_share, chargeback_ratio, merchant_risk_band

## Free vs full version
- Free Kaggle sample: reduced rows, reduced columns, starter notebook, and enough linked finance fraud tables to validate the core workflow.
- Full version: full row volume, richer feature coverage, and extra starter assets for dashboards, SQL, and fraud-analysis work.

## Upgrade to full version
- Full version: https://tekhnikalab.gumroad.com/l/fraud-model-starter-pack
- Upgrade if you need the full linked schema plus starter assets that get you to a fraud dashboard, SQL project, or risk baseline faster.

## Notes
- Contains generated data only and no real personal data.
- Designed as a lightweight free sample for evaluation and discovery.