MovieTime is a Content-Based Movie Recommender System that works on the principles on Unsupervised Machine Learning.
Data of 5000 movies from two separate datasets is preprocessed (this also includes stemming of text data) and EDA is performed. Next, a Bag of Words NLP model is used to vectorize the text input.
The model then uses cosine similarity to offer recommendations. It calculates the cosine similarities of a given movie with all the other movies present in the dataset and returns five movies that are most similar.
The TMDB API is used to fetch the data, Streamlit is used to build the web application, and Heroku is used to deploy it.
MovieTime.mp4
- Numpy
- Pandas
- Scikit Learn
- Nltk
- Streamlit
- Pickle
Hi, I am Arjun Kohli. I'm currently pursuing a Computer Science major at Krea University. Look forward to knowing you!
Anybody interested in contributing to this project is more than welcome to do so!