electricsheepafrica/nigerian-banking-nip-transfers

NIP (Nigeria Inter-Bank Payment) Transfers

Dataset Description

Nigerian Inter-Bank Payment (NIP) transfers with fraud labels

This is a production-grade synthetic dataset with authentic Nigerian banking context, designed for fraud detection, payment anomaly detection, and real-time transaction monitoring.

Key Features

• 🇳🇬 Nigerian Context: Real bank names, states, demographics
• 📊 ML-Ready Labels: Fraud flag (1.0% prevalence)
• ⚡ Large Scale: 3,000,000 rows
• 🎯 Realistic Distributions: Lognormal amounts, temporal patterns, cross-bank transfers
• 📅 Temporal Coverage: 2023-01-01 to 2024-12-31

Dataset Structure

Splits

• Train: 2,100,000 rows (70%)
• Validation: 450,000 rows (15%)
• Test: 450,000 rows (15%)

Schema

Key columns: transfer_id, timestamp, sender/beneficiary accounts & banks, amount_ngn, transfer_type, state, session_id, status, response_time_ms, fraud_flag

Usage

from datasets import load_dataset

# Load full dataset
dataset = load_dataset("electricsheepafrica/nigerian_nip_transfers")

# Load specific split
train_data = load_dataset("electricsheepafrica/nigerian_nip_transfers", split="train")

Labels

• Type: Fraud flag (1.0% prevalence)
• Use Case: fraud detection, payment anomaly detection, and real-time transaction monitoring

Nigerian Context

Includes 14 Nigerian banks (Access, GTBank, UBA, Zenith, etc.), 37 states with population weighting, and authentic NIP transfer patterns including P2P, business, and salary transfers.

Citation

@dataset{nigerian_banking_nip_transfers,
  title = {Nigerian NIP (Nigeria Inter-Bank Payment) Transfers},
  author = {Electric Sheep Africa},
  year = {2025},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/electricsheepafrica/nigerian_nip_transfers}}
}

License

Apache 2.0 - Free for commercial and research use.

Contact

• Organization: Electric Sheep Africa
• Collection: Nigeria Banking & Finance Datasets