Build A Book Recommendation System With Machine Learning [part 1 of 2]
In this video, we'll learn how to build a system to recommend new books for you to read. We'll start by downloading the data, which was scraped from Goodreads. The data files are large (hundreds of millions of rows), so we'll learn some tricks to process this data on your local machine. Then, we'll build a search engine to search the dataset for books that you've read. Finally, we'll use the list of books you've read to find recommendations from others.
Along the way, we'll use pandas, scikit-learn, and numpy. You'll learn about processing large data sets, tf-idf, cosine similarity, and styling pandas dataframe columns.
By the end, you'll have a personalized list of book recommendations, and a project that you can put into your portfolio. In the next part, we'll cover how to improve your recommendations with collaborative filtering.
You can find the project code here - https://github.com/dataquestio/projec... .
And the data can be downloaded at these links:
Data overview - https://sites.google.com/eng.ucsd.edu...
goodreads_books.json.gz - https://drive.google.com/uc?id=1LXpK1...
goodreads_interactions.csv - https://drive.google.com/open?id=1zmy...
book_id_map.csv - https://drive.google.com/uc?id=1CHTAa...
Chapters
00:00 Introduction
00:28 Data and project outline
05:13 Exploring the data in python
09:10 Parsing our book metadata
13:26 Processing book metadata with pandas
18:32 Building a book search engine
27:43 Creating a list of liked books
31:33 Exploring book rating data
34:34 Finding users who like the same books as us
38:08 Finding what books those users liked
43:18 Creating initial book recommendations
44:36 Improving our book recommendations
51:00 Wrap-up and next steps
---------------------------------
Join 1M+ Dataquest learners today!
Master data skills and change your life.
Sign up for free: https://bit.ly/3O8MDef