guptavaibhav35/Netflix_Recommendation

Creating a Netflix Recommendation System with Graph Neural Networks

Description

This project presents a novel approach to analyzing and visualizing the complex network of relationships within a movie dataset, focusing on Netflix titles. It uses a combination of text analysis, machine learning, clustering, and graph theory to uncover insights into the film industry. Key processes include data pre-processing, TF-IDF vectorization for movie descriptions, MiniBatchKMeans clustering to categorize movies, and the construction of a network graph for visualization and recommendation purposes.

Installation Instructions

1. 1.Prepare the Project Directory: Ensure that the dataset and all code files are stored in the same directory.

1. 1.Install Dependencies: Run the following command in your terminal to install all dependencies from requirements.txt:

   pip install -r requirements.txt

1. 1.Run the Analysis Script: Execute netflix_analysis.py to process data and generate models and graphs. The script generates:
2. 2.kmeans_model.joblib: The KMeans clustering model.
3. 3.tfidf_vectorizer.joblib: Vectorized movie descriptions.
4. 4.netflix_graph.pkl: The generated graph.

   python netflix_analysis.py

1. 1.Launch the Streamlit Application: Start the application by running app.py:

   streamlit run app.py

Usage

Run netflix_analysis.py to create the models and graph. Then, launch the Streamlit application using app.py for an interactive experience with the Netflix Recommendation System.

Contact Information

Vaibhav Gupta and Kendelle Grubbs, vaibhav.gupta@vanderbilt.edu, Department of Computer Science, School of Engineering, Vanderbilt University, Nashville, Tennessee 37203 GitHub Repository