Skip to content

luthfiraditya/Ecommerce-Recommendation-System

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Recommender Systems on E-commerce

A recommender system (RS) helps users that have no sufficient competence or time to evaluate the potentially overwhelming, number of alternatives offered by a web site.

–In their simplest form, RSs recommend to their users personalized and ranked lists of items.

The Value of Recommendation System

  • Recommendations are responsible for 70% of the time people spend watching videos on YouTube.
  • 75% of what people are watching on Netflix comes from recommendations, according to McKinsey.
  • 35% of the purchases on Amazon are the result of their recommender system, according to McKinsey.

In this notebook, we will be implementing recommendation model to recommend product to user by using E-commerce data that has 3.265.689 rows and 3 columns applied to 377.716 user and 1.166 product.

Project Overview

Using an ecommerce dataset, this recommendation system was created that recommends to users which product they will like.

Methodology

We will try all the model based that available in surprise library such as SVD, NMF, Slope one etc. And choose the best performance method.

I also using MLflow to tracking the experiment. You can read the full tutorial here

About

I built recommender systems for recommending products to user using Model-based recommendation system.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published