Categorize customers to help merchants to offer promotions or discounts to cardholders.
- Uncover signal in customer loyalty to identify the most 'valuable' individuals.
- Create the right experience for customers. Provide specifically tailored promotions on most relevant merchant.
Data is available on Kaggle: link to Kaggle Competition.
All data is simulated and fictitious, and is not real customer data.
- train.csv - the training set
- test.csv - the test set
- historical_transactions.csv - up to 3 months' worth of historical transactions for each card_id
- merchants.csv - additional information about all merchants / merchant_ids in the dataset.
- new_merchant_transactions.csv - two months' worth of data for each card_id containing ALL purchases that card_id made at merchant_ids that were not visited in the historical data.
Number of Observations: ~ 20 million
Features/ Predicting Variables: Around 30
- Card_id - prime key
- Target_binned - loyalty category: good, bad, neutral
- Month of first purchase
- Authorized_flag - Y: Approved, N: Denied
- Purchase_amount
- State_id
- Merchant_id