🌍 Storyline: The Digital Bazaar

In 2024, e-commerce platforms across Istanbul, Berlin, New York, London, and Paris began noticing strange transaction bursts. Some cards tested with $1 purchases at midnight. Others shipped “gaming accessories” 5,000 km away. Promo codes were being reused from freshly created accounts.

To investigate these global patterns safely, this synthetic dataset recreates realistic fraud behavior across countries, channels, and user profiles — allowing anyone to build, test, and compare fraud-detection models without exposing any real user data.

💡 What makes it special

🧍‍♀️ 6 000 unique users performing ≈300 000 transactions

💳 Multiple transactions per user (40–60) → enables behavioral analysis

🧩 Strong feature correlations — not random noise

🌐 Cross-country dynamics (country, bin_country)

💸 Natural imbalance (~2 % fraud) just like real financial systems

🕓 Time realism — night-time fraud spikes, daily rhythms

🧠 Feature explainability — easy to visualize, model, and interpret

This is a fabricated dataset which is made by merging two dataset, Dataset1.csv and Dataset2.csv .

The final dataset which merged_dataset.csv is a synthetic dataset, using probabilistic imputation to handle missing values.

Balancing the Dataset: The dataset, which was initially imbalanced, was balanced using the ROSE (Random Over-Sampling Examples) package to ensure equal representation of fraudulent and non-fraudulent transactions.

This dataset was used for my group and school project report. You can check out my code for this project, through this https://github.com/slothislazy/DM_AOL

Description

This synthetic dataset, "Fraudulent E-Commerce Transactions," is designed to simulate transaction data from an e-commerce platform with a focus on fraud detection. It contains a variety of features commonly found in transactional data, with additional attributes specifically engineered to support the development and testing of fraud detection algorithms.

Dataset Overview

• Number of Transactions in Version 1: 1,472,952
• Number of Transactions in Version 2: 23,634
• Features: 16
• Fraudulent Transactions: Approximately 5%

Feature Details

1. Transaction ID: A unique identifier for each transaction.
2. Customer ID: A unique identifier for each customer.
3. Transaction Amount: The total amount of money exchanged in the transaction.
4. Transaction Date: The date and time when the transaction took place.
5. Payment Method: The method used to complete the transaction (e.g., credit card, PayPal, etc.).
6. Product Category: The category of the product involved in the transaction.
7. Quantity: The number of products involved in the transaction.
8. Customer Age: The age of the customer making the transaction.
9. Customer Location: The geographical location of the customer.
10. Device Used: The type of device used to make the transaction (e.g., mobile, desktop).
11. IP Address: The IP address of the device used for the transaction.
12. Shipping Address: The address where the product was shipped.
13. Billing Address: The address associated with the payment method.
14. Is Fraudulent: A binary indicator of whether the transaction is fraudulent (1 for fraudulent, 0 for legitimate).
15. Account Age Days: The age of the customer's account in days at the time of the transaction.
16. Transaction Hour: The hour of the day when the transaction occurred.

Purpose

The dataset is intended for use in developing and testing machine learning models for fraud detection in e-commerce transactions. It can also be used for exploratory data analysis, feature engineering, and benchmarking fraud detection algorithms.

Generation Method

The data is entirely synthetic, generated using Python's Faker library and custom logic to simulate realistic transaction patterns and fraudulent scenarios. The dataset is not based on real individuals or transactions and is created for educational and research purposes.

Usage

Feel free to use this dataset for data analysis, machine learning projects, or as a benchmark for fraud detection algorithms. If you use this dataset in your research or projects, please provide proper attribution.

🛡️ Synthetic E-Commerce Fraud & AML Detection Dataset (10k Evaluation Sample)

⚠️ NOTICE: This is a truncated 10,000-row evaluation sample strictly for schema verification and local testing.

💳 [OBTAIN THE 10-MILLION ROW COMMERCIAL LICENSE HERE] > https://buy.stripe.com/8x26oIad4eH9eJf6gJ5wI01

  🚀 Quick Start (Load via Hugging Face)

Data scientists can instantly load this evaluation slice into their Pandas/Python environment using the datasets library: from datasets… See the full description on the dataset page: https://huggingface.co/datasets/apex0data/ecommerce-fraud-detection-synthetic-10k-sampl.

The global e-commerce fraud detection and prevention market was estimated at **** billion U.S. dollars in 2021. Forecasts suggest that this figure will continue to grow steadily in the coming years, surpassing the *** billion dollar mark by 2027.

The dataset contains fake (fraudulent) e-shops data together with legitimate e-shops data. The dataset is balanced and contains 1140 records of 579 fake (fraudulent) and 561 real (legitimate) online shops. Each record contains the following fields: 1. Online shop’s URL; 2. Label - {legitimate, fraudulent}; 3. Domain length - Number of symbols in the host domain name; 4. Top domain length - Number of symbols in the top domain name; 5. Presence of prefix “www” in the active URL of the online shop, values {0 - no, 1 - yes}; 6. Number of digits in the URL; 7. Number of letters in the URL; 8. Number of dots (.) in the URL; 9. Number of hyphens (-) in the URL; 10. Presence of credit card payment, values {0 - no, 1 - yes}; 11. Presence of money back payment, including PayPal, Alipay, Apple Pay, Google Pay, Samsung Pay, and Amazon Pay, values {0 - no, 1 - yes}; 12. Presence of cash on delivery payment, values {0 - no, 1 - yes}; 13. Presence of the ability to use cryptocurrencies for payments, values {0 - no, 1 - yes}; 14. Presence of free contact emails, including Gmail, Hotmail, Outlook, Yahoo Mail, Zoho Mail, ProtonMail, iCloud Mail, GMX Mail, AOL Mail, mail.com, Yandex Mail, Mail2World, or Tutanota, values {0 – email address not found, 1 - free email address, 2 - domain email address, 3 – other email address}; 15. Presence of logo URL, values {0 - no, 1 - yes}; 16. SSL certificate issuer name; 17. SSL certificate expire date; 18. SSL certificate issuer organization name; 19. SSL certificate issuer organization ID, values {1 - Cloudflare, Inc., 2 - Let's Encrypt, 3 - Sectigo Limited, 4 - cPanel, Inc., 5 - GoDaddy.com, Inc., 6 - Amazon, 7 - DigiCert, Inc., 8 - GlobalSign nv-sa, 9 - Google Trust Services LLC, 10 - ZeroSSL, 11 - other organization}; 20. Indication of young domain, registered 400 days ago or later, values {0 - ‘old’ domain name, 1 - ‘young’ domain name, 2 - ‘hidden’}; 21. Domain registration date; 22. Presence of TrustPilot reviews, values {0 - no, 1 - yes}; 23. TrustPilot score, values - real number from 0 to 5 or -1 if no reviews are available; 24. Presence of SiteJabber reviews, values {0 - no, 1 - yes}; 25. Presence in the standard Tranco list, values {0 - no, 1 - yes}; 26. Tranco List rank, values - integer number from 1 to 1000000 or -1 if domain is not listed in the Tranco list.

The size of the E-commerce Fraud Detection market was valued at USD XXX million in 2024 and is projected to reach USD XXX million by 2033, with an expected CAGR of XX% during the forecast period.

The booming e-commerce fraud detection market is projected to reach $45 billion by 2033, driven by AI, ML, and rising online transactions. Learn about key players, market trends, and growth forecasts in this comprehensive analysis.

The e-commerce fraud prevention and detection market is booming, projected to reach $24.63 Billion by 2033. Learn about market drivers, trends, key players like ACI Worldwide and Signifyd, and regional growth in North America and Asia-Pacific. Secure your business with this in-depth analysis.

The booming e-commerce fraud protection software market is projected to reach $15 Billion by 2025, growing at a CAGR of 18%. This comprehensive market analysis explores key drivers, trends, restraints, and leading companies like ClearSale, Riskified, and Signifyd, providing insights for businesses seeking to mitigate online fraud.

The e-commerce fraud protection software market is booming, projected to reach $45 billion by 2033 with a 15% CAGR. Learn about key market trends, leading companies, and the factors driving this explosive growth in online security solutions. Discover how AI and machine learning are transforming fraud prevention.

The e-commerce security fraud prevention market is booming, projected to reach $183.5 billion by 2033 with a 20.2% CAGR. Learn about key drivers, trends, and top players shaping this rapidly evolving landscape. Explore regional market share data & investment opportunities in this crucial sector.

This dataset provides granular records of e-commerce transactions flagged by an AI-powered payment fraud detection system, including transaction details, customer identifiers, risk scores, decision actions, and key model features. It enables fraud prevention teams to benchmark model performance, analyze decision rationale, and optimize fraud detection strategies.

Explore the booming Ecommerce Security Fraud Prevention Software market, projected to exceed $40 billion by 2025 with a 21.1% CAGR. Discover key growth drivers, emerging trends in AI/ML fraud detection, and regional market insights for online businesses.

Explore the booming e-commerce fraud prevention market, valued at $30B in 2025 with a 15% CAGR. Discover key drivers, cloud-based trends, and challenges shaping online security for businesses worldwide.

The booming online fraud detection tools market is projected to reach $45 billion by 2033, driven by soaring e-commerce and AI advancements. Learn about market trends, key players (Kaspersky, LexisNexis, Experian), and future growth projections in this comprehensive analysis.

The size of the Ecommerce Fraud Detection and Prevention Tool market was valued at USD 256.2 million in 2024 and is projected to reach USD 346.32 million by 2033, with an expected CAGR of 4.4 % during the forecast period.

The size of the E-commerce Fraud Prevention and Detection market was valued at USD 11630 million in 2024 and is projected to reach USD XXX million by 2033, with an expected CAGR of XX% during the forecast period.