Synthetic Medical Billing Data With Fraud Labels

10,000 rows of synthetic medical billing data with injected fraud cases. The dataset is designed for fraud detection modeling, though its creator and update date are unknown.

Use Cases

• Classify fraudulent claims using billing codes and patient demographic features.
• Train anomaly detection models on synthetic claim amounts and procedure counts.
• Benchmark fraud prediction models on labeled synthetic provider and patient IDs.

Strengths

• 10,000 synthetic patient billing records.
• Explicitly labeled fraud cases for supervised learning.

Limitations

• Synthetic data may not capture real-world fraud patterns and complexities.
• Unknown column structure limits understanding of available predictive features.

Provenance

Source

Kaggle

Collection Method

Synthetically generated with injected fraud.