Skip to content

sherincheah/amz-ecom-recommender

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Amazon E-Commerce Recommendation System Using Content-Based Filtering

Problem Statement

When shopping on E-commerce platforms, customers may be faced with a bunch of irrelevant recommended products which could possibly divert them out from the shopping platform. Possible reasons include the suggested product price is beyond their budget and item style is too different. Understanding consumer preferences and leveraging on it to retain them in the site could increase the likelihood of them purchasing an item.

Objectives

To recommend users a list of similar products in the same category based on the current product they are browsing.

Installation & Usage

Requirements:

  • Git
  • Python3
$ https://github.com/sherincheah/amz-ecom-recommender.git
$ cd amz-ecom-recommender

$ jupyter notebook

There are 3 notebooks used in this analysis:

Methods

  • Recommendation system part I: Product popularity based system targeted at new customers
  • Recommendation system part II: Model-based collaborative filtering system based on customer's purchase history and ratings provided by other users who bought items similar items
  • Recommendation system part III: When a business is setting up its e-commerce website for the first time without any product rating

cold start reco

Example

result

Further Work

  • Arrange recommended list in ascending order of price so that users will know that is the next best alternative
  • Explore into online supermarket recommendation system e.g. customer browsing for pastas, there will be 2 categories of recommended list - 'similar products' and 'other related products'

Authors

Resources