NETFLIX RECOMMENDATION SYSTEM WITH EXPLORATORY DATA ANALYSIS AND DATA VISUALISATION.

with 3 datasets (here is a kaggle link of all datasets that I've already integrated)

Link for all datasets: https://www.kaggle.com/karan842/netflix-recommendation-and-eda/data

About Project:

This is a recommendation system project based on Netflix dataset from Kaggle, IMDb dataset, Books dataset. Netflix is one the popular online Movies & TV Shows streaming platform. There are thousands of TV Shows and Movies on Netflix those are watched by all over the world. According to Wikipedia net worth of Netflix in 2020 is USD2.761 billion (2020).

Project also contains dataset of BOOKS for finding out how many movies/shows on Netflix are inspired by books📚. We are using two datasets of IMDb data one is IMDb ratings and other is IMDb movies.

In this project I did comparison based Visualization with help of Matplotlib, Seaborn, WordCloud and Plotly. This project is a Natural Language Processing based concept. You can see that all datasets are in the form of text so here I used NLP techniques for making Netflix Recommendation System.

While coding recommendation system I performed two techniques first is TF-IDF(Term Frequency-Inverse Document Frequency) by using this method model performed very well but for accuracy I created other model the second technique which was CountVectorizer and Cosine Similarity.

Approach:

Approach is very simple because like other projects I started with classical way to build Machine Learning approach. Firstly I imported all necessary python libraries and loaded Netflix dataset first and visualize. Then I loaded two datasets of IMDb (ratings and movies). After this I started playing with the collected data i.e. visualization of factors, cleaning the data, making sub dataset and used them for better analysis. The additional part is displaying content with the help of WordCloud.

🟡Recommedation System:

Recommendation System is the main goal of this project as I said earlier that our data is text data, for text data we are using NLP concepts. I did same :p I tested this system on some famous Netflix TV shows and Movies such as F.R.I.E.N.D.S., PEAKY BLINDERS, NARCOS, 3 IDIOTS.

🟣Issue:

I was plotting some visuals with the help of python library called PLOTLY and visuals are not visible on github notebook file. So here are these visuals

Output of code line 98 in notebook file:

Output of code line 136 in notebook file:

🟠Thank You:

Thank you so much, corrections and suggestions are welcome😉