Datasets:

ziadatalabs
/

FreeFraudDetection50M

@@ -1,99 +1,49 @@
----
-license: other
-task_categories:
-  - tabular-classification
-language:
-  - en
-tags:
-  - fraud
-  - fraud-detection
-  - finance
-  - synthetic
-  - tabular
-  - transactions
-size_categories:
-  - 10M<n<100M
-configs:
-  - config_name: default
-    data_files:
-      - split: train
-        path: FreeSyntheticDataFraudDetection50M.parquet
----
-# Zia Data Labs
-## About Zia Data Labs
-We believe great data shouldn't cost a fortune or come with strings attached.
-At Zia Data Labs we do one thing — produce quality datasets and offer them at prices that make sense. No games. No gatekeeping. Just clean, production-grade data you can trust.
-This dataset is completely free. No catch. It's our way of introducing ourselves and letting our work speak for itself.
-We love what we do — and we'd love to earn your business.
-Come find us. More is coming.
-*— The Zia Data Labs Team*
----
-## ⚠️ SAFETY & FIDELITY NOTICE — Zia-Data-Labs
-This dataset is high-fidelity synthetic data engineered to mirror real-world patterns with exceptional accuracy. In benchmark testing, leading AI models treat this data as authentic — recognizing edge cases, flagging anomalies, and generating functional code with zero scrubbing required.
-Because of this realism, improper use during model fine-tuning can trigger deep behavioral shifts in production systems.
-**This dataset is strictly intended for research, evaluation, and development within isolated sandbox or staging environments.**
-Zia-Data-Labs provides all datasets on an "as-is" basis. We do not assume liability for downstream model behavior, deployment risks, or production system impacts. Users are solely responsible for conducting independent safety audits prior to any live deployment.
----
 ## Fraud Detection Dataset
 A production-grade, large-scale dataset containing 50 million transactions designed to serve both the beginner data scientist and the enterprise data engineer.
-What is inside:
 - 50 million rows of high-fidelity synthetic transaction data
 - 19 carefully engineered fields covering core transaction data, velocity signals, geospatial context, device fingerprinting, and diagnostic fraud labels
-- A realistic 2.5% fraud rate simulating the exact needle in a haystack challenge faced by real fraud teams at major financial institutions
-- Granular fraud_type labels enabling multi-class classification, not just binary detection
-File formats provided:
-- sample_50rows.csv — No gate. Inspect immediately.
-- fraud_detection_50M.parquet — Full 50 million row file. No gate. Free access.
-Quick load example:
 import pandas as pd
 df = pd.read_parquet('fraud_detection_50M.parquet')
 print(df.head())
-Schema Overview:
 | Block | Fields |
 |---|---|
-| Core | transaction_id, timestamp, user_id, amount, merchant_id, merchant_category, location_country, payment_method, is_fraud |
-| Velocity & Context | ip_address, device_id, user_age_days, is_first_transaction, time_since_last_txn, auth_result |
-| Diagnostic | fraud_type |
-| Metadata | currency_code, card_network |
-| Branding | data_source |
----
-## Our New Home
 Visit our official new home where our full portfolio of datasets lives and grows:
-Zia-Data-Labs on Hugging Face
-https://huggingface.co/Zia-Data-Labs
----
-## Contact
 zia.data.team@protonmail.com
----
-## License & Usage
-**Zia Data Labs Open Use License**
 This dataset is the exclusive property of Zia Data Labs. It is made freely available for personal research, academic and educational use, and model training and development.

 ## Fraud Detection Dataset
 A production-grade, large-scale dataset containing 50 million transactions designed to serve both the beginner data scientist and the enterprise data engineer.
+**What is inside:**
 - 50 million rows of high-fidelity synthetic transaction data
 - 19 carefully engineered fields covering core transaction data, velocity signals, geospatial context, device fingerprinting, and diagnostic fraud labels
+- A realistic 2.5% fraud rate simulating the exact needle-in-a-haystack challenge faced by real fraud teams at major financial institutions
+- Granular `fraud_type` labels enabling multi-class classification, not just binary detection
+**File formats provided:**
+- `sample_50rows.csv` — No gate. Inspect immediately.
+- `fraud_detection_50M.parquet` — Full 50 million row file. No gate. Free access.
+**Quick load example:**
+```python
 import pandas as pd
 df = pd.read_parquet('fraud_detection_50M.parquet')
 print(df.head())
+```
+**Schema Overview:**
 | Block | Fields |
 |---|---|
+| Core | `transaction_id`, `timestamp`, `user_id`, `amount`, `currency`, `merchant_id`, `merchant_category`, `card_type`, `location_country` |
+| Velocity & Context | `ip_address`, `device_id`, `shipping_zip`, `billing_zip`, `is_international`, `account_age_days`, `prev_transaction_amt` |
+| Diagnostic | `fraud_label`, `fraud_type` |
+| Metadata | `metadata_marker` |
+**Our New Home**
 Visit our official new home where our full portfolio of datasets lives and grows:
+Zia-Data-Labs on Hugging Face — https://huggingface.co/Zia-Data-Labs
+**Contact**
 zia.data.team@protonmail.com
+**License & Usage**
+Zia Data Labs Open Use License
 This dataset is the exclusive property of Zia Data Labs. It is made freely available for personal research, academic and educational use, and model training and development.