Global Customer Invoice Data

Columns

• InvoiceNo: Invoice number. A nominal, 6-digit integral number uniquely assigned to each transaction. Codes starting with 'c' indicate a cancellation.
• StockCode: Product (item) code. A nominal, 5-digit integral number uniquely assigned to each distinct product.
• Description: Product (item) name. Nominal.
• Quantity: The quantities of each product (item) per transaction. Numeric.
• InvoiceDate: Invoice date and time. Numeric. Represents the day and time when a transaction was generated.
• UnitPrice: Unit price. Numeric. Product price per unit in sterling (£).
• CustomerID: Customer number. A nominal, 5-digit integral number uniquely assigned to each customer.
• Country: Country name. Nominal. The name of the country where a customer resides.

Distribution

• Format: CSV (online_retail_II.csv)
• Size: 43.95 MB
• Structure: 8 columns
• Records: Approximately 542,000 valid entries (based on Invoice and StockCode validation counts).

Usage

• Customer Segmentation: Implementing RFM (Recency, Frequency, Monetary) model-based segmentation.
• Profitability Analysis: Predicting customer profitability dynamically over time.
• Market Basket Analysis: Identifying associations between products using sequential pattern mining (prefix and suffix).
• Inventory Management: Forecasting demand for specific stock codes.
• Time Series Analysis: Examining sales trends across different months and years.
• Deep Learning Applications: Testing models such as Bidirectional Deep Recurrent Neural Networks.

Coverage

• Geographic: Covers 38 unique countries, with the United Kingdom representing approximately 91% of the entries.
• Time Range: 1st December 2009 to 9th December 2011.
• Demographic: Customers of a non-store online retail business, primarily including wholesalers and purchasers of gift-ware.

License

Who Can Use It

• Data Scientists: For training and testing classification, clustering, and regression models.
• Retail Analysts: To understand sales performance, cancellation rates, and product popularity.
• Marketing Researchers: For analysing purchasing patterns and customer lifetime value.
• Academic Researchers: Suitable for case studies in data mining and business forecasting.

Dataset Name Suggestions

• UK Online Retail Transactions II
• E-commerce Giftware Sales Records (2009-2011)
• Global Customer Invoice Data
• Non-Store Retail Market Basket Dataset

Attributes