Coupon Recommendation System

About Coupon Recommendation System!

You are hired as a data scientist at a leading shopping mall in the country. The shopping mall has tied up with different restaurants/bars to provide discount coupons to all its customers. The coupons increase the footfalls at these restaurants and help the shopping mall to attract more customers. The organization has been relying upon simple guidelines to determine what coupons are to be provided to the customers, however, the organization feels that they need a more robust model to determine whether a customer will accept the recommended coupon or not to improve the use rate. The organization plans to use a mix of client’s details that they have captured to create this model.

You are provided with the historical data of the recommended coupons along with customer details in the previous years and your task is to come up with a model which would be able to predict whether a customer will accept the recommended coupon.

About the Model

Summary

We developed a machine learning model to predict whether a customer will accept the recommended coupon or not using catboost algorithm.

Algo Applied

This is a binary classification problem to predict the accepted/rejected coupon status. Team has tried Logit, DT, RF, Xgboost and catboost. We got the best results from catboost .

Time Taken to train the model

30 mins , learning rate = 0.1177 for 10K records

Total Entries & Final F Score

24 entries, 0.7714 F1 scores