8000 GitHub - soil55/flannflix: Movie recommender implemented through Flask, with PostGreSQL, SQLalchemy, Dask, Scikit-learn and Altair.
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Movie recommender implemented through Flask, with PostGreSQL, SQLalchemy, Dask, Scikit-learn and Altair.

Notifications You must be signed in to change notification settings

soil55/flannflix

Repository files navigation

flannflix

Movie recommender implemented through Flask, with PostGres, SQLalchemy, Dask, Scikit-learn and Altair.

Features

  • Uses PostGres's full-text search through SQLalchemy to match user-input to films in the database;
  • Collaborative filtering (non-negative factorisation, NMF) with scikit-learn to make recommendations;
  • Distributed handling of ratings matrix with Dask;
  • Flask web-interface;
  • CSS from Bootstrap;
  • Script to generate recommendations for randomised user inputs. Allows insights into which films are recommended most frequently and demonstrates that the same films aren't recommended all the time. The results are displayed in an interactive Altair plot through the web-interface.

Dataset:

Images:

Main Page

Recommendation Page

Chart Page

Requires PostgreSQL.

About

Movie recommender implemented through Flask, with PostGreSQL, SQLalchemy, Dask, Scikit-learn and Altair.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0