/dsc-network-recomendation-systems-lab-staff

Primary LanguageJupyter NotebookOtherNOASSERTION

Recommendation Systems - Lab

Introduction

Now that you've gotten an introduction to collaborative filtering and recommendation systems, it's time to put your skills to test and attempt to build a recommendation system for a real world dataset! For this exercise, you'll be using a dataset regarding the book reviews on the Amazon marketplace. While the previous lesson focused on user-based recommendation systems, you'll apply a parallel process for an item-based recommendation system to recommend similar books at the bottom of the product page.

Objectives

You will be able to:

  • Implement a recommendation system on a real world dataset

Load the Dataset

#Your code here

Load the MetaData

#Your code here

Select Books to Test Your Recommender On

Select a small subset of books that you are interested in generating recommendations for.

#Your code here

Generate Recommendations for a Few Books of Choice

The 'books_data.edgelist' has conveniently already calculated the distance between items for you. Given this preprocessed and data, it's time to employ collaborative filtering to generate recommendations! Generate the top 10 recommendations for each book in the subset you chose. Be sure to print the book name that you are generating recommendations for as well as the name of the books being recommended.

#Your code here

Summary

Well done! In this lab, you effectively created a recommendation system for a real world dataset!