/movieRecommendation

Applied ETL processing on a big data dataset and used content and collaborative algorithms to suggest movies according to user preferences through streamlit webapp.

Primary LanguageJupyter Notebook

Plan of Action

We obtained the open-source Big data movies dataset for this project and used Content-based and Collaborative based recommendations to produce a data frame that can be visualized on a website.

Environment Setup

  1. Install PyCharm to build the front-end of our website.
  2. Jupyter Notebook to develop the model for recommendation and python distribution platforms to build the back-end.
  3. Extract modelled files

Workflow Diagram

image

● Content-based algorithm activity diagram

image

● Cosine similarity

image

References for dataset:

tmdb_5000_credits.csv from Kaggle tmdb_5000_movies.csv from Kaggle Movie_Titles.csv from Grouplens

Kaggle dataset link: https://www.kaggle.com/tmdb/tmdb-movie-metadata?select=tmdb_5000_credits.csv

Group-lens dataset link: https://grouplens.org/datasets/movielens/ framework: https://docs.streamlit.io/

Authors

● @Vishal Solanki ● @Ravleen Kaur ● @Shoaib Ahmed ● @PurveshP0406