Skip to content

BatuhanYilmaz26/TV_Series-Recommender

Repository files navigation

TV_Series-Recommender

Python Jupyter Notebook Streamlit Heroku

About this project

  • Created a dataset of tv shows which has rating greater than 6.0 and has more than 10,000 votes and has been released between 1970-2022 by scraping the IMDb website using Beautiful Soup library and saving the dataframe to a csv file.

  • The dataset contains the following columns:

    • Title: The title of the tv series
    • Year: The year the tv show was released
    • Runtime: The runtime of the tv series
    • Genre: The genre of the tv series
    • Rating: The rating of the tv series
    • Votes: The votes of the tv series
  • Retrieved the tv series trailers for recommended tv series dynamically, using Youtube Data API v3.

  • The recommender system uses the K-Nearest Neighbors algorithm to recommend tv series based on the user's preferences.

  • Built the web app using Streamlit and deployed it on Heroku.