'How to analyze whether a tag on a product influences the buy-decision for a group of customers

Absolute data-science/statistics beginner here, please explain like i'm 15.

I have a database with the following:

  • food-related products with 0-n user-defined tags (such as "vegetarian", "veal", "fish", "low-carb")
  • purchase history of customers (customer_id, product_id, amount, time, ...)

I want to know which tags (or the meaning behind the tags, to be precise, since customers do not actually see the tags) have a measureable impact on the buy-decision of a group of customers (meaning more than 1, less than n). For example, I would imagine that a vegetarian customer would predominantly buy vegetarian products (meaning that yes, there is a group of people for the "vegetarian" tag, where the buy-decision is influenced).

My first idea is that there would be a multimodal distribution in the number of products bought by a customer (because vegetarians would buy more vegetarian products on average), but how would I know to which tag that correlates to if there are multiple?

statisticsdatasetcluster-analysis


Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source

Related Questions