A content based movie recommender system using a cosine similarity
*Here we gonna work on 2 platforms, which are namely: Jupyter Notebook & PyCharm (kindly download the code, given above)
*For Dataset kindly download both the .csv files from the below given link;
https://www.kaggle.com/datasets/tmdb/tmdb-movie-metadata?select=tmdb_5000_movies.csv
https://www.kaggle.com/datasets/tmdb/tmdb-movie-metadata?select=tmdb_5000_credits.csv
*Will also need an API, so for that kindly create one from: https://www.themoviedb.org/
Steps to create API;
-
Once you are landed on tmdb website (https://www.themoviedb.org/)
-
Go to settings & later click on API & create one for yourself.
-
Once done save your respective API key somewhere in your device, will be usefull in further steps!
*Moving forward, open your jupyter notebook & kindly make a movie recommder system folder!
Later add: movie recommender system.ipynb & both .csv files.
Now run the: movie recommender system.ipynb
Once exeuted, you will see a few .pkl files in your movie recommder system folder, don't worry about that it gonna help us to build our website
Now it's time for app.py file which, execute opening pycharm and creating a new project; install streamlit, writing pip install streamlit in the pycharm terminal
Now copy paste all the .pkl file from jypter folder to here
Moving forward change the API key with yours on the 7th line of app.py file
Later type: streamlit run app.py in the pycharm terminal, click on that ip address (Will be similar to: http://192.168.29.41:8501/)
Hurrah!! you have successfully executed movie recommender system; try typing any movies & see the recommended movies related to the movie you have searched for!
Contact: akshendhami@gmail.com, https://www.linkedin.com/in/akshen-dhami22