I am new to Data Science and I want to make Customer Product Analytics for my company(bank). I can have a data of customers, their income, daily transactions, average balance etc and what product(saving certificates etc) they have taken according to their account balance. Can i have a prediction for new customers or existing customers that what product will be suitable for them according to their average balance , income etc? Can the machine learning algorithm predict each product to a particular customer? I got to know cluster analysis and predictive analysis can be useful for such task. But i want to recommend a particular product to a particular customer. Which algorithm can be useful? And from where I have to begin?
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
If you have historic data of earlier purchase by customers. Try to build any classification algo ( Decision tree/ RF) to build rule associated with customer and product. Now you can suggest product to a new customer associating his properties to rules which are already created. Just goole classification algos in R/Python
Clustering ( K means/Hiearchical etc) would be good for customer segmentation, not directly useful for prediction.
https://machinelearningstories.blogspot.com/2017/09/hierarchical-clustering-bottom-up.html
You can also use market Basket analysis for recommending similar products to existing customer.
https://machinelearningstories.blogspot.com/2016/11/recommendation-engine-market-basket.html
-
$\begingroup$ Thankyou for your links, they were really helpful. I will work on it more by using these links. $\endgroup$– MaryamCommented Jun 23, 2018 at 11:24
-
$\begingroup$ @Maryam - please I am interested in learning about the outcome of this project. Thanks $\endgroup$– Z ZCommented Jan 4, 2019 at 9:26