https://github.com/upasanadhameliya/django-movie-recommendor

Movie Website built on python Django framework; Uses Content Based Predictive Model approach to predict similar movies based on the contents/genres similarities
https://github.com/upasanadhameliya/django-movie-recommendor

content-based-filtering content-based-recommendation content-filtering django django-project django3 machine-learning movie-recommendation movielens movielens-dataset python python-3-6 python3

Last synced: about 1 month ago
JSON representation

Movie Website built on python Django framework; Uses Content Based Predictive Model approach to predict similar movies based on the contents/genres similarities

• Host: GitHub
• URL: https://github.com/upasanadhameliya/django-movie-recommendor
• Owner: Upasanadhameliya
• Created: 2020-05-12T09:14:54.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-07-27T11:27:03.000Z (over 4 years ago)
• Last Synced: 2024-10-12T09:21:13.446Z (about 1 month ago)
• Topics: content-based-filtering, content-based-recommendation, content-filtering, django, django-project, django3, machine-learning, movie-recommendation, movielens, movielens-dataset, python, python-3-6, python3
• Language: Python
• Size: 62.8 MB
• Stars: 13
• Watchers: 3
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Django-Movie-Recommendor (Machine Learning)

 Movie Website built on python Django framework.

 Uses ***Content Based Predictive Model*** approach to predict similar movies based on the contents/genres similarities.

 Check out how the website looks on this Link

 



- Libraries used: pandas, numpy, sklearn, imdbpy, pickle

- Python-Django Version: 3.6

- Database: Postgresql

The dataset used for training the model can be downloaded from this link: https://grouplens.org/datasets/movielens/1m/

The model uses batch learning mechanism in contrast to online learning. 

Hence, the dataset will have to be downloaded and the model would have to be trained just once as it would not be updating dynamically.

In case you need to update the movie dataset, the model would have to be retrained after updating the dataset.



##### Run and deploy

1. Clone and download the repository

2. (Do this just once!) Uncomment the three functions in the top level `urls.py` : `imdbscript.get_movies()` , `content_script.get_movies()`, `model_creation.create()`

   - This is only to get the data in the database from the imdb api and to train your model to predict movies.

   - It might take a few hours: approx 6hrs.

   - Comment the functions once again as soon as the website is up else you might get stuck with duplicate

      entries.

      

After following the above steps, you would have the website up and running.