realharshbhatt / movie-recommendation

Recommend movies based on user specified movies, genres and associated people using graph algorithms.

Home Page:http://cinewise.yashrajkakkad.me

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Cinewise - Movie Recommendation Using Graph Theory

PyPI - Python Version PyPI - Django Version PyPI - License

Recommend movies based on user's preferred movies, genres, actors, directors and storywriters. Click here to see it in action.

Preview

This project implements the first two algorithms from a research paper published in IEEE Xplore:

  • Union Colors - a combination of multisource 'breadth first search' (BFS) and union-find data structure.
  • Energy Spread - a variation of multisource breadth first search.

The explanation for these algorithms can be found in the paper. Credits for the same go to the respective authors.

We have created a Django application around the algorithm for a smooth experience. Bootstrap is used for creating a simple yet elegant frontpage.

The Graph

The graph has movie titles, genres and the assoicated directors, actors and story writers as the vertices. These vertices are mutually connected for every movie.

OMDB API

We used OMDB API to obtain movie information and create a graph around it. The data is stored using serialization. (movies.pickle, node.pickle and graph.pickle)

Configuration

You need to manually add the movie attributes in the database for autocompletion. Rest of the steps are the same as any other Django project.

  • Clone this repository
git clone https://github.com/yashrajkakkad/movie-recommendation.git
cd movie-recommendation
  • Install the required packages. Using a virtual environment is recommended.
pip install -r requirements.txt
  • Create a file to store your environment variables for python-decouple to give them at the right places.
nano .env
DB_NAME=''
DB_USER=''
DB_PASSWORD=''
DB_HOST=''
APIKEY=''
SECRET_KEY=''
DEBUG=''
  • If you're not going to use a PostgreSQL database, change the DATABASES variable in mysite/settings.py accordingly. Refer Django documentation for that.

  • If you're going to play with movie related information, you can obtain your API key from OMDB website here. Otherwise keep it blank.

  • DEBUG=True for development and False for production.

  • Make the initial migrations

python manage.py makemigrations
python manage.py migrate

Here's the 'different' part:

  • Open the Django shell:
python manage.py shell
  • Add the vertices:
from utils.Movies import *
from cinewise.models import Node
nodes = load_nodes()
for node in nodes:
    Node(name=node).save()

About

Recommend movies based on user specified movies, genres and associated people using graph algorithms.

http://cinewise.yashrajkakkad.me

License:MIT License


Languages

Language:Python 89.8%Language:HTML 10.2%