I have a Customer Transaction Dataset belonging to an Online Retail Company with the following fields:

DateKey – The date on which the transaction occurred

CustomerKey – Customer ID of the customer who made the transaction

TransactionID – Unique ID to identify a transaction

ProductKey – The product ID of the product sold

ProductName— Name of the product corresponding to the Product ID

ProductSubCategoryName — Name of Sub - Category to which the Product belongs

ProductCategoryName -- Name of Category to which the Product belongs

SalesQuantity – Number of units of the product sold

SalesAmount – Total Amount After Discount (=SalesQuantity X UnitPrice – DiscountAmount)

UnitPrice – Selling price of the product

I need to classify customers based on the above fields as “Very Highly Engaged”, “Highly Engaged”, “Engaged”, “Low Engaged”, “Inactive”

I have derived features such as 1) Frequency of Buying 2) Total Sales Amount 3) Average Number of Days between Transactions 4) Discount Availed 5)Categorical Weight( Amount bought from each category divided by number of transactions in each category which is them summed up) 6) Number of Subcategories and used K-means clustering to divide them into 5 groups.

I am not sure if the above features have yielded a good analysis on the data. This link has a snapshot of the data as well as the graphs obtained. Can anyone tell me what other features I can derive from the data to obtain a better analysis as well as review my current analysis

I have enclosed a snapshot of the data as well the graphs obtained in this link! TIA! Data Link


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.