From analysing their previos transactions how can I predict for what type of product is the customer more likely to take take an EMI

So basically I need a kind of product/ category affinity for EMI for all customers eg - Customer A is more likely to take an EMI on her insurance premium.

One approach I had thought was to broadly categorize the the transactions into 3-4 categories and predict the amount( linear regression) that a customer is going to spend on each of the category in that particular month.eg It is estimated that customer A will spend huge amount on category 1 in month of September, then this customer A will be targetted for a marketing campaign for taking EMI for products in category 1 in September

Another approach I had thought was to do a multi class logistic regression to classify each customer into 3-4 category and marketing can target customers for EMI of specific products

Or I could do clustering and check the conversion for EMI of different types of categories in each cluster

Please let me know if I'm going in the right direction or if there is any other approach I could take or any algorithm I can use. Thank-you

Topic data-science-model marketing regression clustering machine-learning

Category Data Science


This problem could also be formulated as a recommendation system for your marketing department. Given the user's past purchases and the product information and payment methods, you could build a recommendation model to recommend products for your marketing department to push forward for the specific customer.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.